Recombinant-phospholipase A2 production and architecture of inclusion bodies are affected by pH in Escherichia coli

Accepted Manuscript Title: Recombinant-phospholipase A2 production and architecture of inclusion bodies are affected by pH in Escherichia coli Authors: Carlos Calcines-Cruz, Alejandro Olvera, Ricardo M. Castro-Acosta, Guadalupe Zavala, Alejandro Alag´on, Mauricio A. Trujillo-Rold´an, Norma A. Valdez-Cruz PII: DOI: Reference:

S0141-8130(17)33589-4 https://doi.org/10.1016/j.ijbiomac.2017.10.178 BIOMAC 8473

To appear in:

International Journal of Biological Macromolecules

Received date: Revised date: Accepted date:

17-9-2017 30-10-2017 30-10-2017

Please cite this article as: Carlos Calcines-Cruz, Alejandro Olvera, Ricardo M.CastroAcosta, Guadalupe Zavala, Alejandro Alag´on, Mauricio A.Trujillo-Rold´an, Norma A.Valdez-Cruz, Recombinant-phospholipase A2 production and architecture of inclusion bodies are affected by pH in Escherichia coli, International Journal of Biological Macromolecules https://doi.org/10.1016/j.ijbiomac.2017.10.178 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Recombinant-phospholipase A2 production and architecture of inclusion bodies are affected by pH in Escherichia coli Carlos Calcines-Cruz1, Alejandro Olvera2, Ricardo M. Castro-Acosta2, Guadalupe Zavala3, Alejandro

Alagón2,

Mauricio

A.

Trujillo-Roldán1,

and

Norma

A.

Valdez-Cruz1*

[email protected] 1. Programa 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. 2. Departamento de Medicina Molecular y Bioprocesos, Instituto de Biotecnología, Universidad Nacional Autónoma de México. 3. Unidad de Microscopía Electrónica, Instituto de Biotecnología, Universidad Nacional Autónoma de México, Cuernavaca, Mor., México. *Corresponding author: Dra. Norma A. Valdez-Cruz Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México, AP. 70228, México, D.F., CP. 04510, México. Tel.: +52 55 56229192; Fax: +52 55 56223369.

1

ABSTRACT Aggregation of recombinant proteins into inclusion bodies (IBs) is the major drawback of heterologous expression in Escherichia coli. Here, we evaluated the effects of a pH shift after expression induction on recombinant phospholipase A2 production and its aggregation in IBs in E. coli Origami™, as compared to cultures with pH maintained at 7.5 or uncontrolled pH. Cultures shifted from 7.5 to pH 6.5 or 8.5 produced ~15–25% less biomass as compared with those kept at 7.5 or without pH control. The cultures shifted to pH 8.5 showed a ~50% higher yield of acetate per biomass, and the rPLA2 yield was improved 2.4-fold. Purified IBs formed at pH 8.5 containing ~50% of rPLA2, were more susceptible to proteinase-K cleavage and bound less thioflavin-T, indicating lower amyloid content, with the concomitant enrichment of α-helical and random-coil secondary structures, as demonstrated by FTIR. Moreover, only one IB per cell was formed at pH 8.5; instead, more than two were observed under the other culture pH conditions. Nevertheless, under uncontrolled pH conditions, ~300 nm larger IBs were observed. Our work presents evidence of the usefulness of recombinant protein expression cultivated at pH 8.5 allowing the reduction of amyloid content in IBs. Keywords: cultivation pH; acetate; recombinant protein; inclusion body; metabolic overflow; phospholipase A2; E. coli Origami. 1. INTRODUCTION Escherichia coli remains the preferred host for heterologous gene expression at laboratory and industrial scales [1-3]. Nonetheless, due to enhancement of recombinant-protein production and insufficiency of chaperones for folding all proteins properly, aggregation occurs [4]. These aggregates grow into dense particles known as inclusion bodies (IBs), which are the major drawback of E. coli cultures [5,6]. IBs are easily isolated from biomass because of their elevated density, with the additional advantage of being enriched in a recombinant protein [7-9]. Because recombinant proteins are mostly desirable in their monomeric and functional form, they are extracted from IBs in many bioprocesses. On the other hand, the recovery of proteins in a native and soluble state from IBs depends on their cluster characteristics and solubilization conditions. Besides, the refold varies with the biochemical properties of each protein [8]. In fact, the downstream (including refolding) processes are more expensive than upstream ones in bacterial protein production [10,11]. Thus, structural analysis of IBs has become an important field of study. IB architecture determination by infrared spectroscopy, X-ray crystallography, and dye-binding assays has revealed the amyloidal forms, in conjunction with helical and unordered structure [9,12-15]. In addition, some IBs have a

2

catalytic activity indicating the presence of a correctly folded proteins and nativelike variants among intermediates and amyloid aggregates [16-18]. Due to their special architecture and compositional heterogeneity, IBs presents high mechanical stability and sometimes retained biological activity, making them interesting particles, which have been used as biomaterials, catalytic particles, protein storage vehicles, and delivery agents [18-23]. The extraction of native proteins from IBs is often carried out through solubilization with a strong detergent and chaotropic agent [24]. The presence of native or nativelike conformation in the aggregated proteins allows for gentler solubilization methods, which lead to a higher recovery of functional proteins from IBs [8,25]. The proportion of active molecules inside IBs depends on the physical conditions used during protein expression. For example, IBs produced at 25C (referred as “non-classical IBs” [26]) show lower amyloid content and are enriched with active polypeptides as compared to IBs obtained at 37C or 42C, which showed “classical” behavior [27,28]. Our research group has found that recombinant sphingomyelinase D (rSMD) expressed in E. coli BL21(DE3) causes formation of less amyloid IBs when these cells are grown in a bioreactor without pH control, as compared to that maintained at pH 7.5 [29]. During rSMD expression at uncontrolled pH, the medium gradually reached pH 8.5 because of the bacteria own metabolism when cultured in the Super Broth medium. According to these data, we assumed that alkaline pH favors a more relaxed conformation of IBs, which can be easily solubilized [29]. Nevertheless, we could not analyze organic acid production because the culture media used complicated their quantification. Besides, the cultivation strategy did not allow us to test whether acidic pH could yield similar results relative to alkaline pH. On the other hand, the regimen involving a shift to alkaline pH (7.5 to 8.5), has been proposed as a way to diminish the negative effects of acetate on recombinant-protein production in E. coli strain BL21 and to improve the specific growth rate of untransformed E. coli Origami™(DE3) as compared with cultivation at pH 6.5 [30]. Due to such effects as cytotoxicity, hemolysis, neurotoxicity (pre- or post-synaptic), and myotoxicity, phospholipases A2 in snake venom ensure immobilization, killing, and digestion of prey [31,32]. In America, ~85 species of coral snakes have been described [33], with Micrurus laticollaris being one of the most medically relevant species in Mexico [34]. However, due to the modest production of venom by these snakes, few PLA2s have been isolated and characterized [35], and still fewer cloned and expressed. Many snake venom PLA2s share high sequence identity and a conserved structural scaffold, formed principally by an -helix, loops, and a few portions of a -sheet [32,35].

3

The aim of the present work was to assess the effects of a pH shift from 7.5 to 6.5 or 8.5 on rPLA2 expressed in E. coli Origami™ that was induced 2 min after the shift, as compared with these cells grown at pH maintained at 7.5 or without pH control. We evaluated the effects on biomass growth, glucose consumption, recombinant-protein production, organicacid byproducts, and kinetic and stoichiometric parameters. Particularly, structural properties of IBs resulting from rPLA2 overexpression were studied. 2. MATERIALS AND METHODS In this work, all cultures were conducted at least in triplicate in a 1.2-L nominal volume Applikon bioreactor (0.8-L working volume) equipped with temperature, pH, and dissolved oxygen tension (DOT) AppliSens® sensors connected to ADI-1010 Biocontroller (Applikon Biotechnology, Netherlands). To induce recombinant-protein production, isopropyl-β-D-1thiogalactopyranoside (IPTG, 0.1 mM) was added after 5 h of cultivation. The pH regimens analyzed here were as follows: no pH control, pH maintained at 7.5 throughout the culture period, and pH maintained at 7.5 with a subsequent shift to pH 6.5 or pH 8.5, when IPTG was added. 2.1. Strain and cultivation conditions The E. coli Origami(DE3) producer of PLA2 was constructed. Briefly, initially the total RNA was extracted from venom gland of Micrurus laticollaris using the TRIzol reagent (Invitrogen, USA). Then the first-strand cDNA was synthesized with oligo(dT) by reverse transcription using SuperScript III First-Strand Synthesis System (Life Technologies), following the manufacturer’s instructions. Specific PCR was performed using primers to obtain the PLA2 coding sequence (GenBank accession number MF624273), being this gene identical to that used in Valdez-Cruz et al. [9]. This fragment was used as template to add Bam HI and Hind III restriction sites by PCR using specific oligonucleotides. PCR fragment coding for PLA2 with Bam HI /Hind III sites was cloned at pCR2.1-TOPO and amplified in E. coli Xl1blue (Stratagene, La Jolla USA). The fragment coding for PLA2 with restriction sites Bam HI and Hind III was digested by enzymes from New England Biolabs (USA), and then ligated to Bam HI/ Hind III pQE30 expression vector (QIAgen, Germany). Finally, E. coli Origami(DE3) (Novagen, Madison, WI, USA), was transformed with the constructed PLA2 expression vector. The working cell bank was generated by adding glycerol (30% v/v) to conventional shake flasks containing the Luria-Bertani medium (yeast extract 5 g/L, tryptone 10 g/L, and NaCl 10 g/L, pH 7.5) at optical density (OD) at 600 nm (OD600) of 2 absorbance units (AU), and aliquots of 1.0 mL were stored at 70C.

4

All the cultures were grown in a defined medium described by Caspeta et al. [36] with some modifications. The composition (in distilled water; g/L) was as follows: (NH4)2HPO4, 4.0; KH2PO4, 13.3; citric acid, 1.7; MgSO4·7H2O, 1.2; and thiamine hydrochloride, 0.045. Besides, a concentrated solution of trace elements (500) was added at 2 mL/L and had the following composition (g/L): Fe(III) citrate, 100.8; ZnSO42H2O, 22.5; MnCl24H2O, 15.0; EDTA, 14.1; H3BO3, 3.0; CoCl26H2O, 2.5; Na2MoO42H2O, 2.1; and CuCl22H2O, 1.5. The medium was supplemented with 17.5 g/L glucose, 3.0 g/L casamino acids, 0.1 mM CaCl 2, and 50 µg/mL ampicillin. The culture medium, glucose, MgSO4·7H2O, and CaCl2 were autoclaved separately. Thiamine, casamino acids, and ampicillin solutions were sterilized by filtration (0.22 µm, mixed cellulose ester membrane filter, Merck-Millipore, Billerica, MA, USA), and added to the medium just before the inoculum was added. Inocula were grown at 37C and 200 rpm for 12 h in 250-mL shake flasks containing 50 mL of the medium (Duran® Erlenmeyer flask, USA). Bioreactors were inoculated to final OD600 of 0.1 AU and kept at 37C, with air injection at 1 volume of air per volume of culture medium (vvm), and DOT near 30% through an agitation cascade via a proportional-integralderivative (PID) control strategy. The pH level was set to 7.5 after sterilization, and pH was controlled by NaOH or HCl (2.5 N) addition. 2.2. Analytical methods (cell concentration, glucose and organic-acid quantification, and osmolality measurement) Growth of cultures of E. coli Origami™ expressing rPLA2 was followed by OD600 (Spectronic Genesys 20, Thermo, USA). The optical density was converted to dry cell weight (DCW) by means of a linear-correlation standard curve, where 1 AU was equivalent to 0.35 ± 0.02 (g DCW)/L. Culture samples were centrifuged at 7000 × g for 10 min and then filtered (0.2-µm mixed cellulose ester membrane filter, Merck-Millipore, Billerica, MA, USA). The supernatants were used to measure glucose concentration on the Biochemistry Analyzer YSI 2900 (YSI Life Sciences, Yellow Springs, OH, USA) and for quantification of organic acids (acetate, formate, succinate, malate, oxaloacetate, and citrate) by high-performance liquid chromatography (HPLC, Shimadzu, Kyoto, Japan). We employed an Aminex HPX87H column (300 × 7.8 mm; 9-μm internal diameter, Bio-Rad, Hercules, CA, USA), with a mobile phase of H2SO4 (4 mM) at 0.6 mL/min, at 50C with detection by UV absorbance at 215 nm. Organic acid standards were purchased from Bio-Rad (Catalog No. 125-0586). The data were processed in the LC-solution software (Shimadzu, Kyoto, Japan) [37]. Osmolality was measured in filtered supernatants on a Micro-Osmette™ 5004 (Precision Systems Inc., Natick, MA, USA) with automatic Osmometer Sensitivity.

5

2.3. Total cellular protein and recombinant-protein quantification For quantification of total cellular protein, the biomass was recovered by centrifugation (10000 x g for 10 min), resuspended in lysis buffer (50 mM Tris-HCl, 1 mM EDTA, 100 mM NaCl, pH 7.5) with 0.1 mM phenylmethylsulfonyl fluoride (PMSF). The cell suspension was sonicated in a SoniPrep150 (Sanyo-Gallen-Kamp, UK) at amplitude of 10 microns in 10 steps of 30 s at 30-s intervals, on ice. Total cellular protein was treated with solubilization buffer (7 M urea, 2 M thiourea, 2% CHAPS w/v, and 40 mM DTT; all from Sigma-Aldrich, St. Louis, MO, USA) for 30 min. Protein concentrations were determined by the Bradford method (Bio-Rad Protein Assay Kit II, Bio-Rad), and the calibration curve was constructed using bovine serum albumin (BSA) as a standard. OD600 was measured on a 96-well microtiter plate reader (Stat Fax® 4200, Awareness Technology, Inc. Palm City, FL, USA). The samples and standards were prepared in triplicate. The rPLA2 concentration per total cell protein was estimated in at least two samples by densitometry analysis after sodium dodecylsulfate polyacrylamide gel electrophoresis (SDSPAGE) in a 12% gel stained with Coomassie Brilliant Blue R-250, using the Image-Lab™ software on a Gel Doc™ EZ System (Bio-Rad). 2.4. Purification of IBs IBs were isolated as described by Carrió et al. [12] with some modifications. Briefly, biomass pellets were resuspended in lysis buffer (50 mM Tris-HCl, 1 mM EDTA, 100 mM NaCl, pH 7.5) containing protease inhibitor PMSF (0.1 mM). The cell suspension was sonicated on a SoniPrep150 (Sanyo-Gallen-Kamp, UK) at amplitude of 10 microns in 10 steps of 30 s at 30-s intervals, on ice. After that, Nonidet-P40 (Sigma-Aldrich) was added to 1% (v/v), and the samples were agitated for 30 min at 4C. Then, the samples were centrifuged (10000 x g for 10 min) and resuspended in lysis buffer containing 0.5% (v/v) of Triton X-100 (SigmaAldrich) for 30 min. IBs were washed five times with deionized low-conductivity water to remove detergent and DNA traces. The rPLA2 percentage in IBs was estimated in at least three samples by densitometry analysis after SDS-PAGE in a 12% gel, preceded by an IB solubilization step for 30 min (7 M urea, 2 M thiourea, 2% CHAPS w/v, and 40 mM DTT; all from Sigma-Aldrich). 2.5. IB image analysis by transmission electron microscopy (TEM) The protocol for fixing and staining the samples for TEM has been described elsewhere [29]. Briefly, triplicate samples of cells and IBs from different treatment groups were resuspended in 0.16 M sodium cacodylate buffer (pH 7.4) at 4C, and then fixed for 2 h with 4% paraformaldehyde and 2.5% glutaraldehyde in sodium cacodylate buffer pH 7.4 at the same

6

temperature. Next, 1% osmium tetraoxide was used for postfixation of cells or IBs for 90 min at 4C. The samples were rinsed twice in chilled buffer, six times in cold distilled water, and dehydrated in a graded series of ethanol solutions, followed by embedding in Epon. Thin slices were made and then stained with uranyl acetate and lead citrate. A ZEISS Libra 120 plus electron microscope was used to analyze the slices. The percentage of cells bearing IBs was estimated from different micrographs, and a total of 450 cells containing IBs were analyzed. The size of IBs isolated at 10 h after induction was also measured. 2.6. Phospholipase activity The enzymatic activity of the purified IB´s was followed by the hydrolysis of egg yolk phospholipids using a plate test [38]. 2.7. Conformational assessment of isolated IBs IBs resistance to proteolysis was evaluated by enzymatic digestion with proteinase K [29,39]. The aggregates collected at 1 and 10 h after induction were dissolved to OD350 = 1.0 AU in a buffer (50 mM Tris-HCl and 150 mM NaCl, pH 8.0) and proteinase K was added to a final concentration of 50 µg/mL. The solutions were mixed by pipetting, and the initial absorbance (OD350, UV-2450 spectrophotometer, DU 730 Beckman Coulter, USA) was used for normalization. The amyloid content of IBs was assessed by a thioflavin T (Th-T)-binding assay [29,39]. IBs were resuspended to a protein concentration of 50 mg/mL in PBS pH 7.5, mixed with 30 µM Th-T (Sigma-Aldrich, St. Louis MO, USA) and incubated for 30 min at 25C. Excitation wavelength was 445 nm, and emission spectra were recorded between 450 and 600 nm on a spectrofluorometer Luminescence spectrometer LS55 (Perkin Elmer Instruments, USA). Th-T (30 µM) in PBS served as a control. Attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR) was applied to assess the secondary structures of IBs. Shimadzu IRAffinity-1S FTIR spectrometer (Shimadzu, Japan) with a Specac Quest ATR diamond accessory (Specac Limited, England) was used to record the ATR-FTIR spectra of a hydrated thin film of purified IBs in the range 1700–1500 cm-1. A total of 40 interferograms were acquired and averaged. Second-derivative and smoothing procedures with 13 points were applied to all the absorbance spectra. The spectra were normalized to a tyrosine peak at ~1515 cm -1, and the intensities at ~1650 and ~1623 cm-1 were analyzed indicating the abundance of αhelical–random coil and β-sheet structures, respectively [39]. The proteinase K, ATR-FTIR, and Th-T experiments were conducted in triplicate for at least three independent cultures under each condition.

7

3. RESULTS 3.1. Effects of pH on bacterial growth as well as organic-acid production With the intention to demonstrate that pH is one of the main environmental parameter that can direct the aggregation of overexpressed proteins, independently of the expression system and protein [29], we evaluated the expression and the aggregation of rPLA2 in E. coli Origami™ strain (K-12 derivative). This strain has two mutations (thioredoxin reductase, trxB- and glutathione reductase gor-), to increase disulfide bond formation in the cytoplasm [41,42]. Further, a controlled strategy of pH shift was implemented, to allow us to evaluate organic acids and to test acidic or alkaline pH conditions over recombinant protein production and aggregates formation. The cultures with pH shift, in the first five hours were maintained at pH 7.5; after that, pH was shifted to 6.5 or 8.5, and these cultures were compared with the ones maintained at pH 7.5 or without pH control (Fig. 1A). In cultures without pH control, pH decreased gradually to 6.5 after 15 h because of own metabolism of E. coli Origami™-rPLA2 (Fig. 1A). After 8 h of cultivation, the growth was affected by the shifts to pH 6.5 or 8.5 (Fig. 1B), reaching a lower maximal biomass concentration as compared with cultures maintained at pH 7.5 or uncontrolled cultures (Fig. 1B, Table 1). The maximal biomass that was reached when pH was changed to 6.5 was ~35%, ~30%, and ~16% lower than that at uncontrolled pH, pH 7.5, and pH 8.5, respectively. No lag phase was observed in any cultures, nor did we detect significant changes in the specific growth rate (μ) before induction (Table 1). There were no significant differences in the amount of glucose consumed at pH 7.5 and in uncontrolled pH cultures (9.0 ± 0.5 and 8.5 ± 0.3 g/L, respectively), while 9.7 ± 0.1 g/L was consumed at pH 8.5 and only 6.6 ± 0.6 g/L at pH 6.5 (Fig. 1C). Significant differences were observed in biomass/glucose (YX/S) yields, being ~30%, ~20%, and ~10% lower at pH 8.5 than at uncontrolled pH, pH 6.5, and pH 7.5, respectively (Table 1). Furthermore, the glucose specific consumption rate (qs) when pH was shifted to 8.5 was 2.0-, 1.6-, and 1.3-fold higher relative to uncontrolled pH, pH 6.5, and pH 7.5 cultures, respectively (Table 1). In all cultures, DOT was maintained at over 30% to prevent overproduction of organic acids [43]. Acetate and other organic acids were quantified as a measure of metabolites participating in central metabolism and the tricarboxylic acid cycle. Only acetate and succinate were detected. In cultures at pH 6.5, the lowest amount of acetate was detected at the end of cultivation (2.25 ± 0.10 g/L), similar to that detected at pH 7.5 (2.71 ± 0.08 g/L), whereas at pH 8.5 and under uncontrolled conditions, the cells produced 3.63 ± 0.23 g/L and 3.31 ± 0.24 g/L, respectively (Fig. 1D). Particularly, at pH 8.5, the acetate production

8

per biomass unit was ~50% higher as compared with other cultures (Table 1), probably because of the alkali shift [30] and low biomass accumulation. Furthermore, ~0.4 g/L succinate accumulation was observed after 6 h in all cultures (Fig. 1E). In different E. coli strains, changes in pH may modify osmolality thus causing stress [44-46]. In contrast, in our cultures maintained at different pH levels, osmolality was slightly modified in the range of 441 to 544 mOsm/(kg H2O) (Table 1) in comparison with the culture medium alone, which has osmolality of 420 ± 10 mOsmol/kg. Therefore, we can attribute the effects on aggregation, biomass growth [45], and protein accumulation to pH changes. 3.2. Total cellular protein and rPLA2 production improved after the shift to pH 8.5 The influence of the pH regimens on the total protein production and the total protein yield per biomass (Yprot/biom) during culture is shown in Fig. 2A and 2B, respectively. When pH was shifted to 8.5 or maintained at 7.5, total protein accumulation increased by up to 45% at the end of cultivation (10 h of induction). By contrast, in acidic cultures, total protein accumulation remained constant, being lower than 0.6 g/L. Similar Yprot/biom was obtained between the cultures at uncontrolled pH and those maintained at pH 6.5 (0.17 ± 0.04 and 0.25 ± 0.04 gTotal protein/gbiomass, respectively). Nonetheless, 25% more Yprot/biom was obtained at pH maintained at 7.5 (0.31 ± 0.01 gTotal protein/gbiomass). When pH was changed to 8.5, Yprot/biom increased by 57% (0.39 ± 0.02 gtotal protein/gbiomass) as compared with cultures shifted to pH 6.5 (Fig. 2B). Because alkaline pH reduces intracellular acetate accumulation [30], thus reducing its toxicity, this pH allows for higher total protein production per biomass. The accumulation of rPLA2 was quantified by densitometry on gels relative to total cellular protein solubilized and quantified by the Bradford method (Fig. 3A). Accumulation of rPLA2 was higher in cultures at pH 7.5 (0.43 ± 0.06 g/L) and at pH 8.5 (0.40 ± 0.01 g/L) compared with pH 6.5 (0.28 ± 0.06 g/L) or uncontrolled pH (0.23 ± 0.09 g/L) (Table 1). At the end of the cultivation, the same percentage of rPLA2 in purified and solubilized IBs was observed under all conditions: ~50% (Fig. 3B, Table 1). It is worth mentioning that no soluble rPLA2 was detected in all cultures (data not shown). The rPLA2 yield per biomass was higher in cultures maintained at pH 8.5, 7.5, and 6.5 compared with uncontrolled pH (Table 1). These data revealed a 2.4-fold increase in rPLA2 production yield per biomass at pH 8.5 compared with uncontrolled conditions (Table 1). 3.3. Culture pH influences the size and architecture of IBs The IBs inside cells under different pH regimens were visualized by TEM on cross-sections of fixed cells of E. coli Origami™-rPLA2 (Fig. 4). The cells were harvested 1, 5, and 10 h after induction. Under all conditions, the formation of aggregates was favored, even 1 h after

9

induction (Fig. 4), and ~45% of cells contained electrodense aggregates at the end. Moreover, two or more IBs per cell were observed at uncontrolled pH and for pH maintained at 6.5 and 7.5, whereas we observed mostly a single IB per cell when pH was changed to 8.5 (Fig. 4 and 5). This IB content persisted throughout cultivation periods (Fig. 5). This finding suggests that pH affects nucleation during IB formation, taking into account that the formation of a single aggregate was not favored by cultivation at uncontrolled pH, at pH 6.5, and at pH 7.5. The cross-section diameter of IBs obtained after 10 h of induction harvested from different groups was measured in micrographs (Fig. 6A-D). Cumulative distributions of IB diameters showed that the diameter at uncontrolled pH ranged from 100 to 950 nm, as compared with all other conditions, which resulted in diameters between 100 and 650 nm (Fig. 6E). 3.4. Proteolytic susceptibility of IBs at different pH levels To measure the proteolytic susceptibly of IBs and abundance of nonamyloid regions in IBs, proteinase K digestion was carried out at two time points of cultivation (Fig. 7). Differences in the proteinase K induced degradation profile were observed as a function of time; IBs harvested 1 h after induction were more susceptible (~50% were degraded, Fig. 7A), as compared with IBs obtained 10 h post-induction (15% degradation, Fig. 7B). Under uncontrolled conditions and at pH 7.5, IBs were more resistant to proteinase K than IBs formed when pH was shifted to 6.5 or 8.5 (1 h postinduction; Fig. 7A). Moreover, when pH was shifted to 8.5, at the end of cultures (10 h postinduction), IBs were more susceptible to proteinase K digestion as compared with all other treatment groups (Fig. 7B). Furthermore, when pH was shifted to 6.5, intermediate degradation of IBs was observed at 1 h after induction, but 10 h after the induction, the degradation kinetics of IBs were identical to those at pH 7.5. 3.5. IBs isolated from cultures at pH 8.5 have lower amyloid content The amyloid-specific fluorescent dye Th-T, which binds to the -sheet surface of channels structured by “cross-strand ladders” [47,48], was used to characterize the conformation of IBs formed under different pH regimens, 1 and 10 h after induction. As reported elsewhere, Th-T does not bind to precursor polypeptides, monomers, or amorphous aggregates of peptides or proteins [48,49]. Emission maxima for rPLA2-containing IBs were between 472 and 480 nm (Fig. 8). The highest binding of Th-T was observed in IBs produced under acidic conditions (pH 6.5 and uncontrolled pH). Similar behavior was presented by the IBs formed at pH 7.5, but fluorescence intensity was lower (Fig. 8). In fact, the fluorescence signals increased with time. On the other hand, the Th-T-binding assay revealed that IBs from the

10

pH 8.5 group had lower amyloid content at 1 and 10 h after induction, together with a slight increase in Th-T fluorescence (Fig. 8). Additionally, we tested whether the secondary structural elements of the rPLA2-containing IBs were affected by pH regimens, to corroborate the data obtained in proteinase K and ThT experiments. The PLA2 from venom of coral snakes has a primary sequence similar to that of PLA2s with known crystal structure: a general scaffold conformation formed predominantly by a helical protein with loops and turns and a small proportion of -sheets has been predicted elsewhere [31]. Similar content of β-sheets throughout the culture period was observed in IBs collected after 1 and 10 h of induction in all cell groups (Fig. 9A, 9C). Nevertheless, after analyzing three biological replicates, we detected significant differences that indicate a ~20% increase in the amount of -helices in the IBs obtained 10 h after induction when pH was shifted to 8.5 (Fig. 9B). 4. DISCUSSION Here, we present our findings about the production of rPLA2 (from M. laticollaris) in E. coli and how various pH regimens affected the formation of rPLA2-containing IBs and their physicochemical properties. Although it has been mentioned that during expression of a recombinant protein in E. coli, pH cultivation ranging from 6.5 to 7.5 does not affect cell growth and protein expression [30], but some reports indicate that the acidification due to elevated acetate accumulation affects protein synthesis [30,50-52]. For example, when small quantities of acetate accumulate, protein expression is not affected in shake flask E. coli BL21(DE3) cultures expressing glutathione S-transferase (GST), green fluorescent protein (GFP), or cytochrome P450 monooxygenase (CYP) at pH 6.5, 7.5, or 8.5 [30]. By contrast, when exogenous acetate is added (up to 18 g/L), GST, GFP, and CYP accumulation decreases at pH 6.5 [30]. Different strategies have been proposed to diminish the effects of acetate, e.g., mutations [53-55], the use of strains like E. coli BL21 [56-57], and supplementation with amino acids [58]. Particularly, an alkalization shift (to pH 8.5) alleviates growth-inhibitory effects in E. coli, improving the specific growth rate ~3-fold in strains BL21(DE3), Origami(DE3), and DH5, even at a huge concentration of acetate [30]. Here, the pH shift from 7.5 to 8.5 (before induction) caused a decrease in cell growth (by ~18%), as compared with controlled conditions at pH 7.5, but the rPLA2 yield increased more than twofold. In addition, we observed a profound impact of rapid acidification on biomass concentration (a decrease by ~36%) when pH was shifted to 6.5 (from 7.5 before induction) as compared with the uncontrolled pH regimen (Fig. 1B). Furthermore, at pH 8.5, an increase in total protein accumulation was observed at the end of the cultures, which was

11

similar to that observed at pH 7.5. In contrast, under acidic conditions (uncontrolled and pH 6.5), total protein accumulation was lower in comparison with the group where pH was maintained at 7.5. This result suggests that cultivation at pH 8.5 causes the reduction of intracellular acetate [30,50,51], thus allowing for improvement in biomass accumulation of E. coli Origami™-rPLA2 at pH 8.5 compared with pH 6.5. The acid stress response is probably activated after a rapid change from pH 7.5 to 6.5 [59,60], incurring an important metabolic cost and affecting biomass accumulation. Under all the culture conditions, we observed a similar proportion of rPLA2 by densitometry (all in IBs), but as a consequence of the total protein increase, more rPLA2 accumulated at pH 8.5 and 7.5 compared with all other conditions (Table 1). In all the regimens, the accumulation of acetate could be linked to the metabolic overflow caused by the high concentration of glucose at the end of cultivation [61]. The metabolic overflow is caused by a decreased ability of the TCA cycle to metabolize acetate [62,63] with respect to the rate at which glucose was internalized; therefore, pyruvate accumulation causes its transformation into organic acids. In the past, IBs were thought to be aggregates formed by misfolded or unfolded proteins [64]. IB production is becoming increasingly relevant because of its importance in bioprocesses, with IBs serving as a raw material [18,29]. IBs are dynamically formed intracellularly during cultivation of bacteria expressing an exogenous (recombinant) protein [18,29]. Moreover, extracellular stimuli are continuously sensed, resulting in transcriptomic, proteomic and metabolic adjustments in the cell [59,60], and affecting IB formation. Accordingly, temperature and culture duration have been reported to affect aggregation of IBs, as well as their composition and organization [27]. Here, differences in the number and size of IBs inside cells were observed as a function of the pH regimen used (Figs. 5 and 6) and ranged from 100 to 650 nm, where up to ~300 nm larger IBs were formed under uncontrolled conditions compared with controlled ones (Fig. 6). At pH 8.5, most of the cells contained a single IB, while under other conditions two or more IBs per cell were present (Fig. 5). The differences in the IB number and size probably are due to differences in nucleation and IB growth and in the proteins that interact with IBs in response to changes in pH conditions (Fig. 3B). In this respect, it has been reported that genetic deficiencies in DnaK/J, ClpB, or GroEL/S increase the IB number inside cells [65], indicating a change in chaperones interacting with IBs as a function of pH. To determine the effect of pH on the IB scaffold, proteolytic degradation by proteinase K was performed. This enzyme selectively cleaves the peptide bond of aliphatic and aromatic

12

amino acid residues located in hydrophilic domains such as loops and α-helices, without effects on -strands like those in the amyloid core or fibers [18,29,66]. Partial digestion was observed in all IBs, but those formed when pH was shifted to 8.5 revealed greater amounts of proteins extractable by proteinase K (Fig. 7), in agreement with the results obtained by Castellanos-Mendoza et al. [29] who expressed a larger protein under another genetic background. Assuming native folding of rPLA2 owing to higher content of -helices and loops when pH was shifted to 8.5, we tested the phospholipase activity of rPLA2-containing IBs after 1 and 10 h of induction, but those IBs did not have an enzymatic activity. Under acidic conditions, or even at pH 7.5, a smaller proteinase K-sensitive subfraction was observed, suggesting that rPLA2-containing IBs are most likely composed of amyloid structures. In any case, the resistance to proteinase K digestion increased over time, regardless of the culture regimen used. Likewise, the Th-T binding assay indicated amyloidogenic characteristics or -strand content in the rPLA2-containing IBs produced at acidic pH (shifted to 6.5 and uncontrolled pH) or at pH 7.5. By contrast, ~50% less Th-T bound to IBs that formed when pH was shifted to 8.5. Further analysis by ATR-FTIR also showed constant -sheet secondary structure in all IBs but enrichment of -helices in the IBs formed at pH 8.5 (Fig. 9). Previously, we reported that gradual pH change from 7.5 to 8.5 during expression of recombinant sphingomyelinase D in E. coli BL21(DE3) in Super Broth, favors formation of IBs more susceptible to proteolytic degradation and containing less amyloid structures [29]. Thus, differences in IBs produced when pH was shifted to 8.5 may be due to pH stress responses, where the reported proteomic data suggest that cells have specific strategies to reach pH homeostasis [67,71]. These processes (cellular pH stress response and recombinant-protein production) may take place within a similar time frame [70,71]. Among known responses to alkaline stress, E. coli can invert the pH differential between the external and the internal cell membrane by producing acid molecules through amino acid catabolism which neutralizes alkalinity of the cytoplasm, in conjunction with the inward flow of protons through cation/proton antiporters and with a reduction in a cytoplasmic loss of protons [60,68,72-75]. Consequently, we can hypothesize that when pH was shifted to 8.5, less acetate was produced inside the cells, thus improving the metabolism and increasing the total production of protein. This phenomenon must also be accompanied by activation of a heat shock response [69] and concomitant overexpression of rpoH, the master regulator of the heat shock response [59]. The latter change may upregulate chaperones and proteases [69], thereby improving the folding of rPLA2 and resulting in less amyloid conformation inside

13

IBs. Besides, fewer nucleation cores were formed when pH was shifted to 8.5, probably because of larger amounts of active chaperones, given that a lower number of IBs was observed in comparison with the pH 7.5 regimen or acidic conditions. It is important to mention that all observations here discussed were under in vivo conditions in cytoplasm. Thus, the abundance of amyloid structures in those IBs formed under acidic intracellular environments possibly suffer similar effect observed at acidic condition tested in vitro (pH 4) by Peternel et al., [76], where more compact IBs with less extractability characteristics were obtained. 5. CONCLUSIONS Our new results highlight the important role of alkaline pH on the architecture of IBs composed of rPLA2 (14.7 kDa) showing lower amyloid content under different genetic backgrounds, different cytoplasmic redox environment, and different composition of culture media, compared with our previous work [29], expressing another protein of 33.1 KDa (recombinant sphingomyelinase-D). In both studies, we found that alkaline pH has a potential to become a simple alternative way to diminish the -structure in IB´s, that could increase the efficiency of recombinant protein extraction in bioprocesses based on E. coli expressing recombinant protein in aggregates. Furthermore, the enrichment of IBs with nonamyloid protein species by alkaline cultivation pH might also prove advantageous for biomedical applications, like as depots to release of functional protein in vivo [77], increasing the possible uses of IBs and making them more interesting materials. Conflict of interest: All authors declare that they do not have conflict of interest.

14

ACKNOWLEDGEMENTS CCC thanks the scholarship from CONACYT-México. We thank to Eng. Abel BlancasCabrera, M. Sc. Carlos G. Bando-Campos, Chem. Manuel Ortega Hernando, Dr. Alexandra Rodríguez-Sastre and Biol. Lorena López-Griego, for technical assistance in cultures. We specially thank Drs. Laura A. Palomares and Octavio T. Ramírez for allowing the use of the equipment of the Molecular Medicine and Bioprocess laboratory at IBT-UNAM. Also we thank “Unidad de Microscopía Electrónica IBT-UNAM” for TEM equipment rental and image digitalization and analysis support. We thank to Editage (www.editage.com) for English Language editing. This project was developed under the Institutional Program of the “Instituto de Investigaciones Biomédicas-UNAM”: “La producción de biomoléculas de interés biomédico en bacterias y hongos”. Funding: This study was funded by the “Consejo Nacional de Ciencia y Tecnología” CONACYT (220795, 247473, 178528), and “Programa de Apoyo a Proyectos de Investigación e Innovación Tecnológica, Universidad Nacional Autónoma de México” PAPIIT-UNAM (IN-209113, IN-208415). References 1.

N. Ferrer-Miralles, J. Domingo-Espín, J.L. Corchero, E. Vázquez, A. Villaverde: Microbial factories for recombinant pharmaceuticals. Microb Cell Fact 8 (2009) 17. doi: 10.1186/1475-2859-8-17

2.

L. Sánchez-García, L. Martín, R. Mangues, N. Ferrer-Millares, E. Vásquez, A. Villaverde: Recombinant pharmaceuticals from microbial cells: a 2015 update. Microb Cell Fact 15 (2016) 33. doi: 10.1186/s12934-016-0437-3

3.

G. Walsh: Biopharmaceutical benchmarks 2014. Nat Biotechnol 32(10) (2014) 9921002. doi: 10.1038/nbt.3040

4.

M. Martínez-Alonso, E. García-Fruitós, N. Ferrer-Millares, U. Rinas, A. Villaverde: Side effects of chaperone gene co-expression in recombinant protein production. Microb Cell Fact 9 (2010) 64. doi:10.1186/1475-2859-9-64

5.

A.K. Panda: Bioprocessing of therapeutic proteins from the inclusion bodies of Escherichia coli. In: Ghose TK et al. (eds) Biotechnology in India II. Advances in Biochemical Engineering/Biotechnology, vol 85. Springer, Berlin, Heidelberg, (2003) 4393. doi: 10.1007/b11045

6.

W. Wang, S. Nema, D. Teagarden: Protein aggregation--pathways and influencing factors. Int J Pharm 390(2) (2010) 89-99. doi: 10.1016/j.ijpharm.2010.02.025

7.

F. Baneyx: Recombinant protein expression in Escherichia coli. Curr Opin Biotechnol 10(5) (1999) 411-21.

15

8.

A. Singh, V. Upadhyay, A.K. Upadhyay, S.M. Singh, A.K. Panda: Protein recovery from inclusion bodies of Escherichia coli using mild solubilization process. Microb Cell Fac 14 (2015) 41. doi: 10.1186/s12934-015-0222-8

9.

N.A. Valdez-Cruz, G.I. Reynoso-Cereceda, S. Pérez-Rodriguez, S. RestrepoPineda, J. González-Santana, A. Olvera, G. Zavala, A. Alagón, M.A. Trujillo-Roldán: Production of a recombinant phospholipase A2 in Escherichia coli using resonant acoustic mixing that improves oxygen transfer in shake flasks. Microb Cell Fact 16(1) (2017) 129. doi: 10.1186/s12934-017-0746-1.

10.

P. Tucci, V. Veroli, M. Señorale, M. Marín: Escherichia coli: The Leading Model for

the Production of Recombinant Proteins. In: Castro-Sowinski S. (ed) Microbial Models: From Environmental to Industrial Sustainability. Microorganisms for Sustainability, vol 1. Springer, Singapore (2016) 119-147 11.

S. Vemula, R. Thunuguntla, A. Dedaniya, S. Kokkiligadda, C. Palle, S.R. Ronda:

Improved production and characterization of recombinant human granulocyte colony stimulating factor from E. coli under optimized downstream processes. Protein Expr Purif 108 (2015) 62-72. doi: 10.1016/j.pep.2015.01.010. 12.

M. Carrio, R. Cubarsi, A. Villaverde: Fine architecture of bacterial inclusion bodies.

FEBS letters, 471(1) (2000) 7-11. doi: 10.1016/S0014-5793(00)01357-0 13.

M. Carrió, N. González-Montalbán, A. Vera, A. Villaverde, S. Ventura: Amyloid-like

properties of bacterial inclusion bodies. J Mol Biol 347(5) (2005) 1025-1037. doi: 10.1016/j.jmb.2005.02.030 14.

M. Dasari, A. Espargaró, R. Sabaté, J.M. López del Amo, U. Fink, G. Grelle, J.

Bieschke, S. Ventura, B. Reif: Bacterial inclusion bodies of Alzheimer’s Disease βamyloid peptides can be employed to study native-like aggregation intermediate states. ChemBioChem 12(3) (2011) 407-423. doi: 10.1002/cbic.201000602 15.

P. Gatti-Lafranconi, A. Natalello, D. Ami, S.M. Doglia, M. Lotti Concepts and tools to

exploit the potential of bacterial inclusion bodies in protein science and biotechnology. FEBS J 278(14) (2011) 2408-2418. doi: 10.1111/j.1742-4658.2011.08163.x 16.

E. García-Fruitós, N. González-Montalbán, M. Morell, A. Vera, R.M. Ferraz, A. Arís,

S. Ventura, A. Villaverde: Aggregation as bacterial inclusion bodies does not imply inactivation of enzymes and fluorescent proteins. Microb Cell Fact 4 (2005) 27–32. doi: 10.1186/1475-2859-4-27

16

17.

S. Peternel, J. Grdadolnik, V. Gaberc-Porekar, R. Komel: Engineering inclusion

bodies for non denaturing extraction of functional proteins. Microb Cell Fact 7 (2008) 34. doi: 10.1186/1475-2859-7-34 18.

U. Rinas, E. Garcia-Fruitós, J.L. Corchero, E. Vázquez, J. Seras-Franzoso, A.

Villaverde: Bacterial inclusion bodies: discovering their better half. Trends Biochem Sci 42(9) (2017) 726-737. doi: 10.1016/j.tibs.2017.01.005. 19.

C. Díez-Gil, S. Krabbenborg, E. García-Fruitós, E. Vázquez, E. Rodríguez-Carmona,

I. Ratera, N. Ventosa, J. Seras-Franzoso, O. Cano-Garrido, N. Ferrer-Miralles, A. Villaverde, J. Veciana: The nanoscale properties of bacterial inclusion bodies and their effect on mammalian cell proliferation. Biomaterials 31(22) (2010) 5805-5812. doi: 10.1016/j.biomaterials.2010.04.008 20.

E. Hrabárová, L. Achbergerová, J. Nahálka: Insoluble protein applications: the use

of bacterial inclusion bodies as biocatalysts. Methods Mol Biol 1258 (2015) 411-22. doi: 10.1007/978-1-4939-2205-5_24. 21.

J. Seras-Franzoso, K. Peebo, E. García-Fruitós, E. Vázquez, U. Rinas, A. Villaverde:

Improving protein delivery of fibroblast growth factor-2 from bacterial inclusion bodies used as cell culture substrates. Acta Biomat 10(3) (2014) 1354-1359. doi: 10.1016/j.actbio.2013.12.021 22.

J. Seras-Franzoso, A. Sánchez-Chardi, E. García-Fruitós, E. Vázquez, A. Villaverde:

Cellular uptake and intracellular fate of protein releasing bacterial amyloids in mammalian cells. Soft Matter 12 (2016) 3451-3460. doi: 10.1039/C5SM02930A 23.

B. Steinmann, A. Christmann, T. Heiseler, J Fritz, H Kolmar: In vivo enzyme

immobilization by inclusion body display. Appl Environ Microbiol 76(16) (2010) 55635569. doi: 10.1128/AEM.00612-10 24.

F.A.O. Marston, P.A. Lowe, M.Y. Doel, M. Schoemaker, S. White, S. Angal:

Purification of calf prochymosin (rorennin) synthesized in Escherichia coli. Nat Biotechnol 2 (1984) 800-804. doi:10.1038/nbt0984-800 25.

A. Singh, P.K. Panda: Solubilization and refolding of bacterial inclusion body

proteins. J Biosci Bioeng 99(4) (2005) 303-310. doi: 10.1263/jbb.99.303 26.

S. Jevševar, V. Gaberc-Porekar, I. Fonda, B. Podobnik, J. Grdadolnik, V. Menart:

Production of nonclassical inclusion bodies from which correctly folded protein can be extracted. Biotechnol Prog 21(2) (2005) 632-639. doi: 10.1021/bp0497839 27.

N.S. de Groot, S. Ventura: Effect of temperature on protein quality in bacterial

inclusion bodies. FEBS Lett 580 (2006) 6471–6476. doi: 10.1016/j.febslet.2006.10.071

17

28.

A. Vera, N. González-Montalbán, A. Arís, A. Villaverde: The conformational quality

of insoluble recombinant proteins Is enhanced at low growth temperatures. Biotechnol Bioeng 96(6) (2007) 1101-1106. doi: 10.1002/bit.21218 29.

A. Castellanos-Mendoza, R.M. Castro-Acosta, A. Olvera, G. Zavala ,M. Mendoza-

Vera, E. García-Hernández, A. Alagón, M.A. Trujillo-Roldán, N.A. Valdez-Cruz: Influence of pH control in the formation of inclusion bodies during production of recombinant sphingomyelinase-D in Escherichia coli. Microb Cell Fact 13 (2014) 137. doi: 10.1186/s12934-014-0137-9 30.

H. Wang, F. Wang, W. Wang, X. Yao, D. Wei, H. Cheng, Z. Deng: Improving the

expression of recombinant proteins in E. coli BL21 (DE3) under acetate stress: an alkaline

pH

shift

approach.

PLoS

ONE

9(11)

(2014)

e112777.

doi:10.1371/journal.pone.0112777 31.

J.M. Gutiérrez, B. Lomonte: Phospholipases A2: unveiling the secrets of a

functionally versatile group of snake venom toxins. Toxicon 62 (2013) 27e39. doi: 10.1016/j.toxicon.2012.09.006 32.

R.M. Kini: Excitement ahead: structure, function and mechanism of snake venom

phospholipase

A2

enzymes.

Toxicon

42

(2003)

827e840.

doi:

10.1016/j.toxicon.2003.11.002 33.

P. Uetz, J. Goll, J. Hallerman: The TIGR Reptile Database. (2010). Available at:

http://www.reptile-database.org/ (accessed 2017.03.10). 34.

R. Bolaños, L. Cerdas, J.W. Abalos: Venom of coral snakes (Micrurus spp.): report

on a multivalent antivenin for the Americas. Bull Pan Am Health Organ 12(1) (1978) 23– 27. 35.

P. Rey-Suárez, V. Núñez, M. Saldarriaga-Córdoba, B. Lomonte: Primary structures

and partial toxicological characterization of two phospholipases A2 from Micrurus mipartitus and Micrurus dumerilii coral snake venoms. Biochimie 137 (2017) 88-98. doi: 10.1016/j.biochi.2017.03.008 36.

L. Caspeta, A.R. Lara, N.O. Pérez, N. Flores, F. Bolívar, O.T. Ramírez: Enhancing

thermo-induced recombinant protein production in Escherichia coli by temperature oscillations and post-induction nutrient feeding strategies. J Biotechnol. 167(1) (2013) 4755. doi: 10.1016/j.jbiotec.2013.06.001. 37.

G.I. Reynoso-Cereceda, R.I. García-Cabrera, N.A. Valdez-Cruz, M.A. Trujillo-

Roldán: Shaken flasks by resonant acoustic mixing versus orbital mixing: mass transfer

18

coefficient KLa characterization and Escherichia coli cultures comparison. Biochem Eng J. 2016 (2016) 105:379–90. 10.1016/j.bej.2015.10.015. 38.

E. Habermann, K.L. Hardt: A sensitive and specific plate test for the quantitation of

phospholipases. Anal. Biochem. 50 (1972) 163–173. 39.

D. Ami, A. Natalello, G. Taylor, G. Tonon, SM Doglia: Structural analysis of protein

inclusion bodies by Fourier transform infrared microspectroscopy. Biochim Biophys Acta 1764(4) (2006) 793–9. doi.org/10.1016/j.bbapap.2005.12.005. 40.

A.K. Upadhyay, A. Murmu, A. Singh, A.K. Panda: Kinetics of inclusion body

formation and its correlation with the characteristics of protein aggregates in Escherichia coli. PLoS ONE. 2012 (2012) 7:e33951. 10.1371/journal.pone.0033951 41.

M. Berkmen: Production of disulfide-bonded proteins in Escherichia coli. Protein

Expr Purif 82(1) (2012) 240-251. doi: 10.1016/j.pep.2011.10.009 42.

X. Long, Y. Gou, M. Luo, S. Zhang, H. Zhang, L. Bai, S. Wu, Q. He, K. Chen, A.

Huang, J. Zhou, D. Wang: Soluble expression, purification, and characterization of active recombinant human tissue plasminogen activator by auto-induction in E. coli. BMC Biotechnol. 15 (2015) 13. doi: 10.1186/s12896-015-0127-y 43.

M. Losen, B. Frölich, M. Pohl, J. Büchs: Effect of oxygen limitation and medium

composition on Escherichia coli fermentation in shake-flask cultures. Biotechnol Prog 20(4) (2004) 1062-1068. doi: 10.1021/bp034282t 44.

G. Hannig, S.C. Makrides: Strategies for optimizing heterologous protein expression

in Escherichia coli. Trends Biotechnol 16(2) (1998) 54-60. doi: 10.1016/S01677799(97)01155-4 45.

M.T. Jr. Record, E.S. Courtenay, D.S. Cayley, H.J. Guttman: Responses of E. coli to

osmotic stress: large changes in amounts of cytoplasmic solutes and water. Trends Biochem Sci 23 (1998) 143-148. doi: 10.1016/S0968-0004(98)01196-7 46.

A. Weber, S.A. Kögl, K. Jung: Time-dependent proteome alterations under osmotic

stress during aerobic and anaerobic growth in Escherichia coli. J Bacteriol 188(20) (2006) 7165-7175. doi: 10.1128/JB.00508-06 47.

M. Biancalana, K.

Makabe, A. Koide, S. Koide: Aromatic cross-strand ladders

control the structure and stability of beta-rich peptide self-assembly mimics. J Mol Biol 383 (2008) 205–213. 48.

H. LeVine III: Quantification of β-sheet amyloid fibril structures with thioflavin T.

Methods Enzymol 309 (1999) 274-284. doi: 10.1016/S0076-6879(99)09020-5

19

49.

H. Naiki, K. Higuchi, M. Hosokawa, T. Takeda: Fluorometric determination of amyloid

fibrils in vitro using the fluorescent dye thioflavine T. Anal Biochem 177(2) (1989) 244249. doi: 10.1016/0003-2697(89)90046-8 50.

M. De Mey, S. de Maeseneire, W. Soetaert, E. Vandamme: Minimizing acetate

formation in E. coli fermentations. J Ind Microbiol Biotechnol 34(11) (2007) 689–700. doi: 10.1007/s10295-007-0244-2. 51.

A.J. Roe, D. McLaggan, I. Davidson, C. O’Byrne, I.R. Booth: Perturbation of anion

balance during inhibition of growth of Escherichia coli by weak acids. J Bacteriol 180(4) (1998) 767–772. 52.

E.B. Jensen, S. Carlsen: Production of recombinant human growth hormone in

Escherichia coli: Expression of different precursors and physiological effects of glucose, acetate, and salts. Biotechnol Bioeng 36 (1990) 1–11. doi:10.1002/bit.260360102 53.

T.B. Causey, K.T. Shanmugam, L.P. Yomano, L.O. Ingram: Engineering Escherichia

coli for efficient conversion of glucose to pyruvate. Proc Natl Acad Sci USA.101(8) (2004) 2235-40. doi: 10.1073/pnas.0308171100 54.

H. Waegeman, J. Beauprez, H. Moens, J. Maertens, M. de Mey, M.R. Foulquié-

Moreno, J.J. Heijnen, D. Charlier, W. Soetaert: Effect of iclR and arcA knockouts on biomass formation and metabolic fluxes in Escherichia coli K12 and its implications on understanding the metabolism of Escherichia coli BL21(DE3). BMC Microbiol 11 (2011) 70. doi:10.1186/1471-2180-11-70. 55.

H. Waegeman, J. Maertens, J. Beauprez, M. de Mey, W. Soetaert: Effect of iclR and

arcA deletions on physiology and metabolic fluxes in Escherichia coli BL21 (DE3). Biotechnol Lett 34(2) (2012) 329–37. doi:10.1007/ s10529-011-0774-6. 56.

J.N. Phue, J. Shiloach: Transcription levels of key metabolic genes are the cause for

different glucose utilization pathways in E. coli B (BL21) and E. coli K (JM109). J Biotechnol 109 (2004) 21–30. doi: 10.1016/j.jbiotec.2003.10.038. 57.

J. Shiloach, J. Kaufman, A.S. Guillard, R. Fass: Effect of glucose supply strategy on

acetate accumulation, growth, and recombinant protein production by Escherichia coli BL21 (λDE3) and Escherichia coli JM109. Biotechnol Bioeng 49(4) (1996) 421-428. doi: 10.1002/(SICI)1097-0290(19960220)49:4<421::AID-BIT9>3.0.CO;2-R 58.

K. Han, J. Hong, H.C. Lim: Relieving effects of glycine and methionine from acetic

acid inhibition in Escherichia coli fermentation. Biotechnol Bioeng 41 (1993) 316–324. doi: 10.1002/bit.260410305.

20

59.

L.M. Maurer, E. Yohannes, S.S. BonDurant, M. Radmacher, J.L. Slonczewski: pH

regulates genes for flagellar motility, catabolism, and oxidative stress in Escherichia coli K-12. J Bacteriol 187 (2005) 304–319. doi: 10.1128/JB.187.1.304-319.2005. 60.

L.M. Stancik, D.M. Stancik, B. Schmidt, D.M. Barnhart, Y.N. Yoncheva, J.L.

Slonczewski pH-dependent expression of periplasmic proteins and amino acid catabolism in Escherichia coli. J Bacteriol 184(15) (2002) 4246–4258. doi: 10.1128/JB.184.15.4246-4258.2002 61.

G.N. Vemuri, E. Altman, D.P. Sangurdekar, A.B. Khodursky, M.A. Eiteman: Overflow

metabolism in Escherichia coli during steady-state growth: transcriptional regulation and effect of the redox ratio. Appl Environ Microbiol 72(5) (2006) 3653-3661. doi: 10.1128/AEM.72.5.3653-3661.2006 62.

C. Wittmann, J. Weber, E. Betiku, J. Krömera, D. Böhmb, U. Rinas: Response of

fluxome and metabolome to temperature-induced recombinant protein synthesis in Escherichia coli. J Biotechnol 132 (2007) 375-384. doi: 10.1016/j.jbiotec.2007.07.495 63.

A.J. Wolfe: The acetate switch. Microbiol Mol Biol Rev 69(1) (2005) 12-50. doi:

10.1128/MMBR.69.1.12-50.2005 64.

F. Baneyx, M. Mujacic: Recombinant protein folding and misfolding in Escherichia

coli. Nat Biotechnol 22(11) (2004) 1399-408. doi: 10.1038/nbt1029 65.

M. Carrio, A. Villaverde: Role of molecular chaperones in inclusion body formation.

FEBS Lett 537 (2003) 215-221. doi: 10.1016/S0014-5793(03)00126-1 66.

M. Morell, R. Bravo, A. Espargaró, X. Sisquella, F.X. Avilés, X. Fernàndez-Busquets,

S. Ventura: Inclusion bodies: specificity in their aggregation process and amyloid-like structure.

Biochim

Biophys

Acta

1783

(2008)

1815–1825.

doi:

10.1016/j.bbamcr.2008.06.007 67.

C. Bordi, L. Théraulaz, V. Méjean, C. Jourlin-Castelli: Anticipating an alkaline stress

through the Tor phosphorelay system in Escherichia coli. Mol Microbiol 48(1) (2003) 211– 223. doi: 10.1046/j.1365-2958.2003.03428.x 68.

E. Padan, E. Bibi, M. Ito, T.A. Krulwich: Alkaline pH homeostasis in bacteria: new

insights.

Biochim

Biophys

Acta

1717(2)

(2005)

67–88.

doi:

10.1016/j.bbamem.2005.09.010 69.

D. Taglicht, E. Padan, A.B. Oppenheim, S. Schuldiner: An alkaline shift induces the

heat shock response in Escherichia coli. J Bacteriol 169(2) (1987) 885–887. doi: 10.1128/jb.169.2.885-887.1987

21

70.

J.C. Wilks, J.L. Slonczewski: pH of the cytoplasm and periplasm of Escherichia coli:

rapid measurement by green fluorescent protein fluorimetry. J Bacteriol 189 (2007) 5601– 5607. doi: 10.1128/JB.00615-07 71.

J. Yu, J. Xiao, X. Ren, K. Lao, X.S. Xie: Probing gene expression in live cells, one

protein

molecule

at

a

time.

Science

311

(2006)

1600–1603.

doi:

10.1126/science.1119623 72.

E.T. Hayes, J.C. Wilks, P. Sanfilippo, E. Yohannes, D.P. Tate, B.D. Jones, M.D.

Radmacher, S.S. BonDurant, J.L. Slonczewski: Oxygen limitation modulates pH regulation of catabolism and hydrogenases, multidrug transporters, and envelope composition in Escherichia coli K-12. BMC Microbiol 6 (2006) 89. doi: 10.1186/14712180-6-89 73.

H. Saito, H. Kobayashi: Bacterial responses to alkaline stress. Sci Prog 86 (2003)

277–282. doi: 10.3184/003685003783238635 74.

J.L. Slonczewski, B.P. Rosen, J.R. Alger, R.M. Macnab: pH homeostasis in

Escherichia

coli:

measurement

by

31P

nuclear

magnetic

resonance

of

methylphosphonate and phosphate. Proc Natl Acad Sci USA, 78 (1981) 6271–6275. doi: 10.1073/pnas.78.10.6271 75.

E. Yohannes, D.M. Barnhart, J.L. Slonczewski: pH-dependent catabolic protein

expression during anaerobic growth of Escherichia coli K-12. J Bacteriol 186(1) (2004) 192–199. doi: 10.1128/JB.186.1.192-199.2004 76. Š. Peternel, S. Jevševar, M. Bele, V. Gaberc-Porekar, V. Menart: New properties of inclusion bodies with implications for biotechnology. Biotechnol Appl Biochem (2008) 49, 239–246. doi:10.1042/BA20070140 77. M.V. Céspedes, Y. Fernández, U. Unzueta, R. Mendoza, J. Seras-Franzoso, A. Sánchez-Chardi, P. Álamo, V. Toledo-Rubio, N. Ferrer-Millares, E. Vásquez, S. Schwartz Jr., I. Abasolo, J.L. Cochero, R. Mangues, A. Villaverde: Bacterial mimetics of endocrine secretory granules as immobilized in vivo depots for functional protein drugs. Scientific Reports, 6:35765 (2016). doi: 10.1038/srep35765

22

23

Figure captions Fig. 1. Kinetics of E. coli Origami™-rPLA2 in cultures carried out without pH control (filled circles), with constant pH at 7.5 (filled triangles) and shifted from 7.5 to 6.5 (open circles) or 8.5 (open triangles) prior to rPLA2 induction. Data show the mean and standard deviation for at least three biological replicates per pH condition. The dashed line shows the start of IPTG induction (5 h of culture), just after pH change. Profiles of pH in 1.2 L bioreactor cultures (A), biomass growth kinetics (B), glucose consumption (C), acetate (D) and succinate (E) production of E. coli Origami™-rPLA2 cultures. Fig. 2. Total cellular protein production analysis in E. coli Origami™-rPLA2 cultures in bioreactors under different pH strategies at 1 (black), 3 (horizontal striped), 5 (vertical striped) and 10 h (gray) after induction (cultures carried out without pH control, with constant pH at 7.5 and shifted from 7.5 to 6.5 or 8.5 after rPLA2 induction). Volumetric production of total cellular protein (A) and total cellular protein yield (gtotal protein/gBiomass) (B). The mean and standard deviation for three biological replicates per pH condition are shown. Fig. 3. Analysis of total cellular (A) and inclusion bodies (B) protein content by SDS-PAGE (12%) under reducing conditions at the end of E. coli Origami™-rPLA2 (14.7 kDa) cultures under different pH conditions (cultures carried out without pH control -UC-, with constant pH at 7.5 and shifted from 7.5 to 6.5 or 8.5 prior to rPLA2 induction). M: molecular mass standards (kDa). Fig. 4. Cross-sections examination of E. coli Origami™-rPLA2 by transmission electron microscope (TEM). Micrographs show cells collected at 1, 5 and 10 h after induction under different pH conditions: cultures carried out without pH control (A), with constant pH at 7.5 (C) and shifted from 7.5 to 6.5 (B) or 8.5 (D) after rPLA2 induction. Scale bar is 1 µm. Fig. 5. Percentage of cells containing IBs. One (black bar), two (light gray bar) and three and more (dark gray) IBs at 1, 5 and 10 h after induction in different pH conditions (cultures carried out without pH control, with constant pH at 7.5 and shifted from 7.5 to 6.5 or 8.5 after rPLA2 induction). A total of 450 cells presenting IBs were counted for this comparison. Fig. 6. Transmission electron micrographs of IBs obtained from cultures carried out without pH control (A), with constant pH at 7.5 (C) and shifted from 7.5 to 6.5 (B) or 8.5 (D) after rPLA2 induction. Bar represents 2 m. Cumulative frequency of IB diameters (E) measured in transmission micrographs at the end of cultures without pH control (filled circles), with constant pH at 7.5 (filled triangles) and shifted from 7.5 to 6.5 (open circles) or 8.5 (open triangles) after rPLA2 induction. At least 400 IBs were measured for each condition.

24

Fig. 7. Proteinase-K digestion profiles of rPLA2 IBs collected at 1 h (A) and 10 h (B) after induction from cultures carried out without pH control (filled circles), with constant pH at 7.5 (filled triangles) and shifted from 7.5 to 6.5 (open circles) or 8.5 (open triangles) after rPLA2 induction. Data are the mean and standard deviation of at least three biological replicates. Fig. 8. Th-T binding properties of rPLA2 inclusion bodies collected at 1 h (A) and 10 h (B) after induction from cultures carried out without pH control, with constant pH at 7.5 and shifted from 7.5 to 6.5 or 8.5. The fluorescence of Th-T alone is shown as a control. The characteristic spectrum of at least three biological replicates is presented. Fig. 9. ATR-FTIR structural analysis of rPLA2 IBs harvested at the end of cultures. Second derivative spectra (A) of absorbance normalized by the tyrosine peak at ~1515 cm-1, the peaks employed to assess the α-helix and random coil is ~1654 cm-1 and β-sheet are indicated at ~1623 cm-1, from pH 6.5 (dashed line), pH 7.5 (dotted line), 8.5 (dashed-dotted line) and without pH control (continuous line). Heights of peaks at ~1654 cm-1 (B) and ~1623 cm-1 (C) of rPLA2 IBs harvested 1 h (black bar) and 10 h (white bar) after induction. Data show the mean and standard deviation of three biological replicates.

25

26 Figr-1

Figr-2

27

Figr-3

28

Figr-4

29

Figr-5

30

Figr-6

31

Figr-7

32

Figr-8

33

34 Figr-9

Table I: Stoichiometric and kinetic parameters of growth and rPLA2 production for E. coli Origami™ growing at the pH conditions assayed. The mean and standard deviation for at least three biological replicates per pH condition are shown. uncontrolled pH µ (h-1)

0.589 ± 0.030

pH 6.5

pH 7.5

pH 8.5

0.588 ± 0.034 0.572 ± 0.035 0.595 ± 0.007

Xmax (g)

3.30 ± 0.16

2.24 ± 0.19

3.00 ± 0.07

2.57 ± 0.05

YX/S (gBiomass/gGlucose)

0.46 ± 0.03

0.41 ± 0.05

0.39 ± 0.01

0.31 ± 0.03

qS (gGlucose/gBiomass*h)

1.4 ± 0.2

1.7 ± 0.3

2.1 ± 0.3

2.8 ± 0.5

YAcetate/X (gAcetate/gBiomass)

1.04 ± 0.04

1.06 ± 0.07

0.99 ± 0.10

1.50 ± 0.13

YAcetate/S (gAcetate/gGlucose)

0.36 ± 0.01

0.33 ± 0.03

0.28 ± 0.03

0.38 ± 0.05

[PTS] (g/L)

0.50 ± 0.17

0.54 ± 0.04

0.88 ± 0.10

0.95 ± 0.01

46 ± 7

53 ± 3

49 ± 6

42 ± 5

[rPLA2] (g/L)

0.23 ± 0.09

0.28 ± 0.06

0.43 ± 0.06

0.40 ± 0.01

YrPLA2/X (grPLA2/gBiomass)

0.07 ± 0.02

0.13 ± 0.02

0.15 ± 0.02

0.17 ± 0.03

rPLA2 in IBs (%)

35