Cry1Ac toxin and protoxin affect intestinal barrier function by altering zonula occludens-1 and activating mitogen-activated protein kinase pathways in a Caco-2 cell monolayer model of the human intestinal mucosa

doi:10.21203/rs.3.rs-3256816/v1

Download PDF

Research Article

Cry1Ac toxin and protoxin affect intestinal barrier function by altering zonula occludens-1 and activating mitogen-activated protein kinase pathways in a Caco-2 cell monolayer model of the human intestinal mucosa

https://doi.org/10.21203/rs.3.rs-3256816/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The Cry proteins produced by Bacillus thuringiensis are widely used as bioinsecticides to control crop plagues; they have been considered innocuous for vertebrates. Currently, distinct truncated forms of the genes encoding Cry toxins are expressed in genetically transformed crops that are used for animal and human consumption. Cry1Ac is one of these proteins to which humans are exposed. Our group has demonstrated that Cry1Ac protoxin and toxin are mucosal and systemic immunogens that can activate macrophages via mitogen-activated protein kinase (MAPK) pathways, while the protoxin is also a promising adjuvant that can improve tumour immunity and provide protection in different mouse infection models. Moreover, the intragastric administration of Cry1Ac toxin provokes moderate allergenicity and lymphoid hyperplasia in the large intestine. Therefore, we characterised the intestinal epithelial effects after stimulation with Cry1Ac. We used an in vitro model with the Caco-2 C2BBe1 cell line that mimics a gut monolayer with brush border proteins and well established expression of tight junction proteins. We evaluated whether Cry1Ac toxin and protoxin could activate MAPK signalling pathways and whether this activation modified the expression of zonula occludens 1 (ZO-1), a tight junction protein. Cry1Ac activated extracellular signal-regulated kinase (ERK), p38 and c-Jun N-terminal kinase (JNK); upregulated IL-8; altered permeability; and dyregulated ZO-1. Finally, we identified several Cry1Ac binding proteins in the intestinal epithelium that could be involved in these effects.

Epithelial Cells

Permeability

ZO-1

Cry1Ac

MAPK

Cry proteins of Bacillus thuringiensis have been widely used to control insects and are commonly used as agricultural biopesticides. The insecticidal mechanism of Cry proteins involves several steps after the insects have ingested the protoxin: it is solubilised and proteolytically processed into the active form (toxin) in the midgut (Bravo, Gill et al. 2007). The spores containing Cry proteins have been applied over crop fields by aspersion to combat pests and have been considered safe. Currently, distinct truncated forms of the genes encoding Cry toxins are expressed by transgenic crops that are consumed by animals and humans (Gould 1998). These transgenic crops have become a primary method to control pests and to enhance agricultural production, and their use continues to expand throughout the world. Although the reported expression levels of these modified Cry toxins in plants have been reported to be very low, they are known to show wide variation within the distinct parts of the plants and between distinct specimens collected within a culture field, so the amount of Cry toxins contained in plants and the amount consumed by animals and people might not be estimated precisely (Rasmussen, Goulding et al. 1998). Several studies have confirmed the safety of genetically modified (GM) crops (Zhang, Zhuo et al. 2014, Szymczyk, Szczurek et al. 2018, Zhang, Hou et al. 2021). However, the immunological effects and the potential effects on the intestinal barrier function in mammals have not been evaluated, so the potential risk of long-term GM consumption is still unknown. Thus, there is a demand for additional evidence sustaining GM food safety.

Most studies evaluating the safety of Cry proteins in mammals have consisted of feeding animals with transgenic crops and then assessing parameters such as body weight, food consumption, histopathology and blood chemistry. Researchers have confirmed the safety of these proteins because there were no abnormalities in any of the measured parameters (Zhang, Zhuo et al. 2014). However, previous studies that have evaluated the immunological effects of some Cry1A proteins in experimental mice have demonstrated their mucosal and systemic immunogenicity (Moreno-Fierros, Garcia et al. 2000) (Guerrero, Dean et al. 2004) and the adjuvant effects(Gonzalez-Gonzalez, Garcia-Hernandez et al. 2015) and immunoactivation properties in macrophages,(Torres-Martinez, Rubio-Infante et al. 2016) (Rubio-Infante, Ilhuicatzi-Alvarado et al. 2018)dendritic cells and lymphocytes(Ibarra-Moreno, Ilhuicatzi-Alvarado et al. 2021).

The results of our previous studies support that Cry1Ac toxin and protoxin are not innocuous to mammals. Indeed, Cry1Ac protoxin has been proposed as a systemic and mucosal adjuvant because it can increase protection in several infection models (Vazquez, Moreno-Fierros et al. 1999, Carrasco-Yepez, Rojas-Hernandez et al. 2010) and also serves to improve tumour immunity(Servin-Garrido, Ilhuicatzi-Alvarado et al. 2022). Cry1Ac toxin and protoxin also are potent immunogens and both can activate macrophages by inducing upregulation of costimulatory molecules (CD80 and CD86); pro-inflammatory cytokines like interleukin 6 (IL-6), monocyte chemoattractant protein 1 (MCP-1) and tumour necrosis factor alpha (TNF-a) (Moreno-Fierros, Garcia-Hernandez et al. 2013); and mitogen-activated protein kinase (MAPK) signalling pathways (Torres-Martinez, Rubio-Infante et al. 2016). Although several Cry1Ac protoxin binding proteins have been identified in macrophages, until now only cell surface heat shock protein 70 (HSP70) have been related to some of the Cry1Ac-induced activation effects. Moreover, some studies have suggested that Cry1Ac might affect the intestinal mucosa: (Vazquez-Padron, Gonzales-Cabrera et al. 2000) found that Cry1Ac protoxin induces transient hyperpolarisation of mouse jejunum and binds to some intestinal proteins, although they did not identify these proteins. Moreover, (Santos-Vigil, Ilhuicatzi-Alvarado et al. 2018) evaluated the allergenicity of Cry1Ac toxin by oral administration in mice. They demonstrated that Cry1Ac toxin is moderately allergenic and leads to lymphocyte hyperplasia in the large intestine. These findings increase the need for more studies testing the effects of Cry proteins at the intestinal level.

Until now, the evaluation of activation effects of Cry1Ac on mammals has mainly been focused on immune cells such as macrophages (Szymczyk, Szczurek et al. 2018). In the present study, we aimed to determine whether Cry1Ac could also stimulate intestinal epithelial cells. For this purpose, we used Caco-2 C2BBe1 cells, a human colon carcinoma cell line that has the ability to differentiate spontaneously into a monolayer and that have functional characteristics of mature intestinal enterocytes (Sinnecker, Ramaker et al. 2014). We evaluated the effect of simulation with Cry1Ac toxin and protoxin on the activation of MAPKs, cytokine production and the barrier function of Caco-2 cell monolayers. Finally, to identify potential proteins that might be related with the Cry1Ac-induced activation of epithelial cells, we performed precipitation assays with Cry1Ac toxin and protoxin using membrane fractions isolated from Caco-2 cells. We identified binding proteins by matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) tandem mass spectrometry (MS/MS).

Recombinant Cry1Ac

Recombinant Cry1Ac protoxin was purified from Escherichia coli JM103 (Pos900) cultures as described previously (Rojas-Hernandez, Rodriguez-Monroy et al. 2004). Cry1Ac protoxin was solubilised and activation with trypsin was carried out as described previously (Guerrero, Dean et al. 2004). Cry1Ac toxin purity was analysed by sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie blue staining. Endotoxin contamination in the purified Cry1Ac toxin was tested using the E-TOXATE kit (Sigma-Aldrich, St. Louis, MO, USA, ET0200-1KT) following the manufacturer’s instructions; the levels were below 0.1 IU/ml. The purified proteins were passed through a polymyxin B resin column (Sigma-Aldrich, P1411) to remove any possible endotoxin remnants.

Cell culture

Colorectal adenocarcinoma epithelial Caco-2 (clone C2BBe1) cells were obtained from ATCC (Manassas, VA, USA) and cultured according to the supplier’s instructions. Confluent Caco-2 cell monolayers (about 14–20 days) were treated with 40 µg/ml of purified Cry1Ac toxin or protoxin or 10 µg/ml lipopolysaccharide (LPS; Sigma-Aldrich No. 12117850). The treatment was very short term (2.5–60 min or 2 h) longer at 2, 24, 48 or 72 h, depending on the experiment. The positive controls were 2 µM ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA, Invitrogen™ E1219) or 5 µg cholera toxin (Sigma-Aldrich No.C8052); the treatment lasted for 2 or 24 h. The LPS and EGTA concentrations were chosen based on previous studied: they cause an inflammatory response (Szymczyk, Szczurek et al. 2018).

MAPK activation

To evaluate whether Cry1Ac promotes MAPK activation, 1 × 10 ⁶ Caco-2 cells were seeded per well in a 6-well plate with 1 ml of medium supplemented with 10% foetal bovine serum (FBS) for at least 15 days, until confluent monolayers were visible under the microscope. Each well was stimulated with 40 µg/ml of Cry1Ac toxin or protoxin or 10 µg/ml of LPS (the positive activation control). The Cry1Ac concentration for these assays was chosen based on a dose-response curve assay evaluating cytokine production and metabolic activity determined by the MTT assay for Caco-2 cells; the tested doses were from 2.5 to 80 µg/ml. These assays revealed that 40 µg/ml was the optimal Cry1Ac dose to induce MAPK activation and cytokine production. The LPS concentration used was chosen based on a previous report in which the activation of macrophages was analysed (Torres-Martinez, Rubio-Infante et al. 2016). Untreated cells served as the negative control.

To evaluate MAPK activation, supernatants were collected at 2.5, 5, 15, 30 or 60 min after exposure and stored at -70°C until use. Immediately after the cells were washed with phosphate-buffered saline (PBS), they were recovered with a cell scraper and centrifuged at 400 g for 5 min at 4°C. The pellets were suspended in lysis buffer (50 mM HEPES pH 7.4 250 mM NaCl, 5 mM ethylenediaminetetraacetic acid [EDTA], 0.1% Nonidet P-40, 1 mM phenylmethylsulphonyl fluoride [PMSF], 0.5 mM dithiothreitol [DTT], 5 mM sodium fluoride, 0.5 mM sodium orthovanadate, 5 mg/mL aprotinin and 5mg/mL leupeptin) and incubated for 30 min at 4°C.The lysates were centrifuged at 400 g for 5 min (4°C) (Torres-Martinez, Rubio-Infante et al. 2016). Next, 40 µg of protein was separated by SDS-PAGE. The separated protein was transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were blocked for 1 h at room temperature with 5% non-fat milk in Tris-buffered saline with 0.05% Tween-20 (TBS-T). Then, the membrane was incubated with primary antibody diluted 1:1000 in TBS-T for 24 h. The primary antibodies were against p44/42 (No. 137F5 Cell Signaling, Danvers, Massachusetts, United States), phosphorylated p44/42 (p-p44/42 No.T202/Y204, Cell Signaling), p38 (No. 9212S, Cell Signaling), p-p38 (No. T180/Y182,Cell Signaling), c-Jun N-terminal kinase (JNK)(No. 56G8, Cell Signaling) and p-JNK (No. T183/Y185, Cell Signaling). The membranes were washed with TBS-T and incubated with the secondary antibody, anti-rabbit IgG-IRDye800 diluted 1:10000, for x 1h. The fluorescent immunoreactive bands were detected with an Odyssey CLx scanner (LI-COR, Lincoln, NE, USA) at 800 nm.

The same blot was used to measure the levels of phosphorylated and non-phosphorylated MAPKs. First, the membrane was incubated with the primary antibody against the phosphorylated MAPK as described above. Then, the membranes were stripped by incubating twice with 25 mM glycine and 1% SDS (pH 2 ) for 15 min, blocked and incubated with the primary antibody against the non-phosphorylated MAPK. The same procedure described above was used to visualise the fluorescent immunoreactive bands.

Transepithelial electrical resistance (TER) measurement

Caco-2 cells (6 × 10⁵) were seeded on 0.4-µm pore size trans-well filters (Corning-Costar, No. 431222, Acton, MA, USA). TER was measured using a Milicell-Electrical Resistance System (Milicell-ERS) MERS00001. TER is presented as Ω × cm² or as a relative value normalised to control cells. Culture medium was exchanged every 2 days until the monolayers achieved 800 Ω of electrical resistance, which indicates that monolayers were completely formed. At that time, Cry1Ac toxin or protoxin or the positive control was added. After the stimulus, TER was measured at 2 and 24 h. After 24 h, the paracellular flux assay was performed. First, the medium was replaced with fresh medium and the cells were incubated for 24 h. TER was measured again but without the stimulus to determine whether it returned to the initial value.

Paracellular flux assay

After treatment with Cry1Ac toxin or protoxin or the positive control for 24 h, monolayers were rinsed three times with pre-warmed Hanks’ Balanced Salt Solution (HBSS) supplemented with 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), glucose and Ca²⁺/Mg²⁺. Then, 100 µg of fluorescein isothiocyanate (FITC)-dextran (4 kDa, No. 53379 Sigma-Aldrich) was added to the apical side and incubated for 2 h at 37°C. Medium from the bottom chamber was collected, and the amount of diffused dextran was measured using a fluorometer. Emission values were normalised to control cells (set to 100%).

Cytokine measurements

The supernatants of the untreated cells and cells treated with 40 µg/m Cyr1Ac toxin or protoxin or 10 µg/ml LPS for 24 h at 37°C were recovered. The levels of TNF, IL-2, 1L-4, 1, IL-6, IL-10, IL-17 and interferon (IFN γ) were measured using an enzyme-linked immunosorbent assay (ELISA) kit or a CBA Human Inflammation Kit (BD Bioscience, No. 551811 Franklin Lake, NJ, USA) according to the manufacturer’s instructions with the following modifications: all steps were performed in microtubes, and the standard curve was started at a concentration of 0.625 pg/mL to determine concentrations below the curve recommended by the manufacturer. Fluorescence intensities were measured using a FACS Calibur BD cytometer and CBBDA software.( Becton Dickinson, Lomas de Chapultepec Ciudad de México, Mexico.)

Confocal microscopy

For confocal microscopy, 1 × 10 ⁶ Caco-2 cells were seeded in Millicell EZ slide (No. C86024), with 1 ml of supplemented medium and 10% FBS. The cells were cultured for at least 15 days and when confluent monolayers were visible under the microscope, the cells were fixed with 4% paraformaldehyde (PFA) for 10 min at room temperature and permeabilised with 0.2% Triton X-100 in PBS for 5 min at room temperature. The EZ slides were then blocked for 20 min in PBS containing 3% bovine serum albumin. The slides were incubated with an anti-zonula occludens 1 (ZO-1) antibody (Invitrogen, MA, USA, No. 33-9100, lot. VC299822) was incubated overnight at 4°C. Then, the EZ slides were washed and incubated with species-specific fluorescently labelled secondary antibodies (goat anti-rabbit conjugated to Alexa Fluor 488; Invitrogen, A11008) for 1 h at room temperature. Coverslips were mounted in Prolong medium containing 4’,6-diamidino-2-phenylindole (DAPI; Thermo Fisher Scientific No. P36935) and analysed using a confocal laser microscope (Leica TCS SPE).

Membrane protein extraction

Caco-2 plasma membrane proteins were purified to identify proteins that might interact with Cry1Ac toxin and protoxin. Approximately 6 × 10⁷ Caco-2 cells were used per assay. Cells were collected with a cell scraper and washed twice with PBS. After that, the Plasma Membrane Protein Extraction Kit (Abcam, Cambridge, UK, ab65400) was used to isolate the membrane proteins, following the manufacturer’s instructions.

The plasma membrane proteins isolated from Caco-2 cells were incubated for 2 h at 4°C with Sepharose beads covalently coupled to Cry1Ac toxin or protoxin. Unbound membrane proteins were washed out and then separated with SDS-PAGE. The gel was stained with Coomassie blue and then submitted to MALDI-TOF MS/MS to identity the Caco-2 membrane proteins that precipitated with Sepharose coupled to Cry1Ac toxin and protoxin.

All samples were processed by the Proteomic Unit Service at the Biotechnology Institute of National Autonomous University of Mexico (UNAM). The samples were reduced with dithiothreitol (Sigma-Aldrich, No. D5545), alkalised with iodoacetamide (Sigma-Aldrich, I6125) and digested ‘in gel’ with Promega sequencing-grade modified trypsin (V5280). For digestion, 50 mM ammonium bicarbonate (pH 8.2) was used and the samples were incubated for 18 h at 37°C. The resulting peptides were desalted with Zip-tips with 0.6 µl of C18 resin (Merck Millipore, Burlington, Massachusetts, No. ZTC18S096). The samples were sequenced in the University of California Davis, Proteomics Core Facility, Genome Center.

Protein identification

Tandem mass spectra were obtained with the Orbitrap fusion Lumos Tribrid and Dionex UltiMate 3000 RSLC Nano System Flow (Thermo Scientific). The Proteome Discoverer 1.4.1.14 software was used to identity the proteins with the searching parameters of Sequest H. Scaffold 5.1.2 visualisation software (Proteome Software Inc., Portland, OR, USA) Sequent (version IseNode in Proteome Discoverer 1.4.1.14; Thermo Scientific) and X! Tandem (The GPM, thegpm.org; version X! Tandem Alanine [2017.2.1.4]) were used to analyse the tandem mass spectra. Sequest was set up to search uniprot-homo-2022.08.29–18.04.40.32.fasta (200515 entries) assuming the digestion enzyme trypsin. X! Tandem was set up to search reverse concatenated uniprot-download_true_format_fasta_query_HOMO_20SAPIENS-2022.08.22–20.13.13.23 database (unknown version, 3345178 entries), also assuming trypsin digestion. Sequest and X! Tandem were searched with a fragment ion mass tolerance of 0.60 Da and a parent ion tolerance of 20 ppm. Carbamidomethyl of cysteine was specified in Sequest and X! Tandem as a fixed modification. Deamidation of asparagine and glutamine and oxidation of methionine were specified in Sequest as variable modifications. Glu → pyro-Glu of the N-terminus, ammonia loss of the N-terminus, Gln → pyro-Glu of the N-terminus, deamidation of asparagine and glutamine and oxidation of methionine were specified in X! Tandem as variable modifications.

Scaffold_5.2.0 (Proteome Software Inc.) was used to validate the MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established at > 18.0% probability to achieve a false discovery rate (FDR) of < 5.0% by the Percolator posterior error probability calculation (Käll et al. 2008). Protein identifications were accepted if they could be established at greater than 99.0% probability and contained at least two identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm (Nesvizhskii et al. 2003). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins were annotated with Gene Ontology (GO) terms from the National Center for Biotechnology Information (downloaded 4 September 2022) (Ashburner et al. 2000).

Cry1Ac toxin and protoxin activate ERK1/2 and p38 in Caco-2 cells

To determine whether Cry1Ac toxin and protoxin activate ERK1/2, JNK and p38, we stimulated Caco-2 cells with 40 µg/ml Cry1Ac toxin or protoxin or 10 µg/ml LPS and collected their lysates at different time points to analyse protein expression with western blotting. Both Cry1Ac toxin and protoxin induced ERK1/2 phosphorylation (Fig. 1), but they showed differences. For Cry1Ac protoxin, the maximum activation was at 2.5 min and then decreased over time. On the other hand, Cry1Ac toxin activated ERK1/2, but soon after, at 5 min, the activation also declined. LPS led to stronger activation of ERK1/2: it started to increase after 2.5 min and was sustained for at least 60 min. Cry1Ac toxin and protoxin both triggered JNK phosphorylation (Fig. 1). JNK activation was stronger and faster in cells treated with Cry1Ac protoxin compared with cells treated with LPS. Cry1Ac toxin also activated JNK and the activation last at least 60 min. The kinetics of p38 activation induced by Cry1Ac toxin and protoxin and LPS were strikingly different (Fig. 1). p38 was not activated until 15 min after the addition of Cry1Ac protoxin, but it was activated 2.5 min after the addition of Cry1Ac toxin. In the case of LPS, activation occurred within 5 min and then decreased. Thus, the activation of MAPKs after stimulation with LPS or Cry1Ac exhibited distinct kinetics. These signalling pathways have been related to tight junction remodelling, leading in some cases to enhanced permeability of the monolayer or triggering cytokine production (Gonzalez-Mariscal, Tapia et al. 2008, Aggarwal, Suzuki et al. 2011).

Cry1Ac toxin and protoxin induce IL-8 production

Treatment of Caco-2 cells with Cry1Ac toxin or protoxin (40 µg/ml) or LPS (10 µg/ml) induced the secretion of IL-8 at 24 h. Cry1Ac protoxin led to the greatest production of IL-8, around 280 pg/ml. We also noted slight IL-6 production at 24 h after treatment with Cry 1Ac toxin (data not shown).

Cry1Ac increases the epithelial permeability and disrupts tight junctions

To determine whether Cry1Ac-induced activation of MAPKs and cytokine production in Caco-2 cells is related to any functional effect, we assessed the integrity of Caco-2 cell monolayers after treatment with Cry1Ac toxin or protoxin, or LPS or EGTA (positive controls). Interestingly, all of the treatments caused a significant decrease in TER. The effect of Cry1Ac protoxin was greater that the effect of Cry1Ac toxin and the positive controls (Fig. 3a). We also evaluated the macromolecular permeability using fluorometric detection of 4 kDa FITC-dextran. We found that the paracellular flux was increased twofold in monolayers treated with Cry1Ac toxin or protoxin compared with untreated monolayers (Fig. 3B).

We then evaluated whether the reduction in TER was associated with an alteration of tight junction–related proteins, specifically ZO-1. We used confocal microscopy to assess the change in ZO-1 expression in Caco-2 cell monolayers (Fig. 3C). Quantification of the pixel intensity of ZO-1 revealed a significant Cry1Ac-induced decrease in ZO-1 (Fig. 3D). Apical ZO-1 localisation (red fluorescence) was not preserved. At the sites of cell contacts, the ZO-1 pattern changed from strictly linear in untreated cells to wavy in Cry1Ac toxin– and Cry1Ac protoxin–treated cells. Such a wavy pattern has been reported to be correlated with increased epithelial permeability (Chanez-Paredes 2022). We did not observe this wavy pattern in positive controls treatments because all the cell–cell junctions had been lost (Fig. 3C).

Identification of plasma membrane proteins to which Cry1Ac toxin and protoxin bind

We precipitated plasma membrane proteins with Cry1Ac toxin and protoxin and then submitted them to SDS-PAGE. Images of the Coomassie blue–stained gels are shown in Fig. 4. There are evident differences in the pattern of precipitated bands between Cry1Ac protoxin and Cry1Ac toxin. There is a bold band on 70 kDa that was precipitated with Cry1Ac protoxin (Fig. 4a) but not with Cry1Ac toxin. Instead, there is a band around 50 kDa that was precipitated with Cry1Ac toxin (Fig. 4b).

Following MALDI-TOF MS/MS analysis of the membrane proteins, we identified 1686 in the precipitations with Cry1Ac toxin and protoxin; however, only a few candidates had at least two unique peptides. To establish strict criteria to find potential Cry1Ac receptors that might be responsible for the biological effects observed in epithelial cells, we filtered the results by the criteria of having at least two unique peptides and > 25% coverage. Moreover, we chose only proteins that have been reported to be localised in the plasma membrane. After this filtering, only 54 proteins remained that had precipitated with Cry1Ac protoxin (Table 1) and 39 proteins that had precipitated with Cry1Ac toxin (Table 2). We noted that 42 of those 54 proteins have an extracellular region, including the transferrin receptor; galectin 3; annexin 4; Voltage-Dependent Anion Channel (VDAC1) ; the clathrin receptor; and HSP71, HSP90 and HSP120. It is important to mention that cell-surface HSP70 has been previously identified as a potential receptor partially responsible of the Cry1Ac protoxin–induced activation effects in macrophages.

We then used even stricter filtering, selecting proteins with more specific receptor-related functions: evidence of being located in the plasma membrane, having extracellular region and having functions related to biological adhesion or immune properties. These criteria reduced the list to six proteins with a function in biological adhesion – CD2-associated protein, a chloride intracellular channel, cystatin, demoplakin, filamin-A and S100 – and five proteins related to the immune response – HSP60, dermcidin, filamin-A, galetin-3, protein disulphide isomerase and serotransferrin (Fig. 5). The complete list of 1686 identified proteins are presented in the supplementary figures.

Table 1 Caco-2 plasma membrane proteins that precipitated with Cry1Ac protoxin

Table 2 Caco-2 plasma membrane proteins that precipitated with Cry1Ac toxin

The results of the present study indicate that Cry1Ac toxin and protoxin can provoke moderate alterations in TER and paracellular flux and can activate MAPKs and IL-8 secretion in well-differentiated Caco-2 cell monolayers. We observed some differential effects between Cry1Ac toxin and protoxin: in general, the toxin induced more pronounced alteration of the barrier function than the protoxin. The implications of these effects are relevant as the Cry1Ac toxin is expressed in GM crops used for human consumption.

We tested the effects of Cry1Ac toxin and protoxin in Caco-2 cell monolayers because they exhibit functional characteristics of mature intestinal enterocytes. They differentiate and form tight junctions between cells, and thus serve as a model of gut permeability, integrity and metabolic routes activities (Engle, Goetz et al. 1998). The effects of Cry1Ac proteins on barrier function are important, as the normal function of the gastrointestinal tract depends on the maintenance of the mucosal barrier to prevent pathogenic microorganisms from escaping the intestinal lumen to invade the host. The Cry1Ac-induced change in the junctional protein ZO-1 may explain the functional alterations, as the intestinal physical barrier is maintained because the epithelial cells form a monolayer via intercellular connections such as adherents and tight junctions that provide specific permeability to different substances and molecules (Meng, Klingensmith et al. 2017). However, further investigation is required to determine the potential effects of Cry1Ac toxin and protoxin on additional junctional proteins.

The activation of MAPK signalling pathways in gut epithelial cells is associated with a reduction in the expression of tight junction–related proteins, which then alters the barrier function and the intercellular connections (Lei, Cheng et al. 2014, Li, Bai et al. 2021). We demonstrated that both Cry1Ac toxin and protoxin activated the ERK1/2 pathway within 2.5 min and also reduced the expression of ZO-1, decreased TER and increased paracellular flux. Based on these findings, it is tempting to speculate that Cry1Ac alters the gut barrier function by activating MAPK signalling pathways. Consistently, it has been reported that ERK1/2 activation in well-established cell monolayers is associated with destabilisation of tight junctions due to the inhibition of claudin-2 (Aggarwal, Suzuki et al. 2011). Because claudin-2 and ZO-1 are major components of tight junctions, the normal expression of these proteins makes a ‘tight’ epithelium, and the lack or decreased expression of them provokes a ‘loose’ epithelium. Moreover, tight junctions have been recognised as paracellular diffusion barriers that regulate the flux of small solutes and ions because they form interactions with adjacent cells (Wu, Huang et al. 2019).

Although we measured the activation of MAPKs by western blotting using whole cell lysates, it is important to point out that ERK1/2 is expressed surrounding intercellular junctions and its activation leads to the disruption of adherens and tight junctions (Aggarwal, Suzuki et al. 2011). So, it is possible that ERK1/2 activation in Caco-2 cells mediates the altered ZO-1 expression observed in the cells treated with Cry1Ac toxin and protoxin. These changes might lead to the increased permeability of the monolayer and decreased TER. Increased paracellular flux due to barrier disruption has been also associated with activation of the JNK signalling pathway (Samak, Narayanan et al. 2011, Samak, Gangwar et al. 2014, Li, Bai et al. 2021). Consistently, we observed that Cry1A toxin and protoxin also activated JNK. This effect may contribute to disrupt the barrier function. While Cry1Ac protoxin activated JNK more robustly than Cry1Ac toxin, the former might have only a minor effect on the barrier function. This results lead us to propose that other factors besides MAPKs modulate the permeability of cell monolayers treated with Cry1Ac toxin or protoxin.

Activation of the p38 signalling pathway is known to be related to cytokine production in well-established epithelial monolayers (Khan, Kang et al. 2004, Liang and Kitts 2018). In Caco-2 cells treated with Cry1Ac toxin and protoxin, the p38 signalling pathway was activated continuously for 60 min and also prompted the upregulation of IL-8. Surprisingly, this effect was greater for Cry1Ac toxin and protoxin than for LPS. It is important to mention that IL-8 is a major neutrophil chemoattractant and its expression is also augmented in inflamed mucosa (Mitsuyama, Toyonaga et al. 1994). Additional in vivo studies are needed to determine whether Cry1Ac toxin and protoxin could lead to neutrophil recruitment as a response to an inflamed mucosa and whether this recruitment might also be implicated in decreased TER and increased permeability. The fact that Cry1Ac protoxin provoked the greatest IL-8 secretion raises additional questions: does this effect lead to potential neutrophil recruitment? Does this effect indicate the adjuvant capacity of Cry1Ac protoxin?

Although Cry1Ac decreased TER and increased the permeability ratio, the changes were much less pronounced than those provoked by cholera toxin or EGTA. These effects are relevant because we are exposed to Cry1Ac by consuming GM plants. The outcomes suggest that Cry1Ac toxin and protoxin disrupt the intestinal barrier due to ERK and JNK activation and loss of the ZO-1 protein. Of note, this disruption does not destroy the epithelial junctions as EGTA and cholera toxin do. Nevertheless, further investigation is required to elucidate the details of the mechanism of these effects.

We identified numerous potential Cry1Ac receptors in Caco-2 plasma membranes that precipitated with Cry1Ac toxin and protoxin. Therefore, besides the coverage and unique peptides criteria, we used strict filtering criteria to reduce the list and to select more reliable candidate receptors. We selected only proteins with described localisation in plasma membrane and a role in biological functions such as adhesion and immune-related functions. These criteria reduced the list to six proteins with adhesion functions – CD2-associated protein, a chloride intracellular channel, cystatin, demoplakin, filamin-A and S100 – and five proteins with immune functions – HSP60, dermcidin, filamin-A, galetin-3, protein disulphide isomerase and serotransferrin. Additional studies are needed to confirm these findings, but there is evidence that supports their potential participation in the Cry1Ac-induced effects in Caco-2 cells.

HSP60 has been reported as a receptor for Listeria adhesion protein in Caco-2 cells (Wampler, Kim et al. 2004), so is tempting to propose that HSP60 could interact with Cry1Ac toxin and drive some of the cellular processes, such as activation of MAPKs, that we observed in this study. The interaction of Cry1Ac protoxin and the identification of binding proteins has been reported before in macrophages; interestingly, HSPs were among them. Indeed, HSP70 acts as a receptor of Cry1Ac protoxin in RAW264 cells, a macrophage cell line, as it partially mediates some of the Cry1Ac-induced effects such as the activation of ERK1, the upregulation of MCP1 and the recruitment of macrophages to the peritoneal cavity (Rubio-Infante, Ilhuicatzi-Alvarado et al. 2018). Additional inhibition studies are required to explore the participation of HSP60 in the Cry1Ac-induced effects in Caco-2 cells.

Serotransferrin could be linked to Cry1Ac-induced proinflammatory effects (Harel, Rubinstein et al. 2011) because its expression is linked to upregulation of proinflammatory cytokines in enterocytes. This finding is consistent with the fact that Cry1Ac toxin and protoxin prompted IL-8 release in Caco-2 cells. IL-8 has been reported to stimulate proinflammatory cytokines in macrophages.

Galectin 3 has been implicated in numerous adhesion and immune processes including in the chemoattractant properties for macrophages and monocytes (Sano, Hsu et al. 2000). The proteomics results confirm and complement the Cry1Ac-induced effects we observed. We identified proteins involved in the cellular activities that are triggered by Cry1Ac protoxin and toxin such as proinflammatory cytokine production and activation of MAPK signalling pathways.

Other Cry1Ac-induced effects we detected were the changes in electrical resistance and alteration of the paracellular flux. Desmoplakin and S-100 have been reported to be downregulated in epithelial dysplasia (Gupta, Nitoiu et al. 2015). Perhaps Cry1Ac toxin and protoxin interact with cystatin to enhance the loss of availability that leads the membrane disruption.

Filamin-A and CD2-associated protein may also be involved as Cry1Ac receptors in this system. ERK1/2 activation could be triggered by cytoskeletal alterations and filamin-A could modulate this activation (Zhao, Ma et al. 2016). On the other hand, CD2-associated protein has been associated with p38 activation (Kalland, Oberprieler et al. 2011). Although we identified several Caco-2 membrane proteins based on the Cry1Ac toxin and protoxin immunoprecipitation assays, the functional correlation of this interaction needs to be corroborated in additional studies. Nevertheless, the theoretical analysis give us clues and indicates that the effects of Cry1Ac toxin and protoxin are due to a complex of proteins that lead to different responses in the epithelial cells.

Although no lethal effects have been reported in animal models using Cry1Ac toxin and protoxin, the results of the present study show that Cry1Ac toxin and protoxin stimulation in well-established Caco-2 monolayers provokes effects that may be related to gut inflammation. These proteins enhanced the permeability of the Caco-2 cell monolayer, increased IL-8 production and activated MAPKs. Moreover, the induction of these effects may be mediated through receptors located in the plasma membrane, but additional studies are required to probe their functional role on Cry1Ac induced effects on epithelial cells.

Ethical Approval

The project was approved by ethical committee of the Facultad de Estudios Superiores Iztacala UNAM ( project approval no 1484), but non animals or human were used for the data presented in this study.

Competing interests

The authors have no relevant financial or non-financial interests to disclose

Authors' contributions

This work was conceived and designed by ISBJ and LMF. Cell line culture, permeability assays protocols were guided by MS, Signalling experiments and membrane protein experiments were guided by DAI. The experiments, methodology, validation, investigation, data analysis, figure preparation, wrote draft to final manuscript version ISBJ. Finally LMF was in charge of investigation, methodology, validation, writing—review and editing draft to final manuscript version, project administration, and funding acquisition. All authors reviewed the manuscript.

Funding

This work was supported by PAPIIT IN202923 and PAPCA 2022-20. Authors LMF, DIA and ISBJ has received research support from PAPIIT and PAPCA.

MS declare that no funds, grants, or other support were received during the preparation of this manuscript

Availability of data and materials

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Aggarwal, S., T. Suzuki, W. L. Taylor, A. Bhargava and R. K. Rao (2011). "Contrasting effects of ERK on tight junction integrity in differentiated and under-differentiated Caco-2 cell monolayers." Biochem J 433(1): 51-63.
Bravo, A., S. S. Gill and M. Soberon (2007). "Mode of action of Bacillus thuringiensis Cry and Cyt toxins and their potential for insect control." Toxicon 49(4): 423-435.
Carrasco-Yepez, M., S. Rojas-Hernandez, M. A. Rodriguez-Monroy, L. I. Terrazas and L. Moreno-Fierros (2010). "Protection against Naegleria fowleri infection in mice immunized with Cry1Ac plus amoebic lysates is dependent on the STAT6 Th2 response." Parasite Immunol 32(9-10): 664-670.
Engle, M. J., G. S. Goetz and D. H. Alpers (1998). "Caco-2 cells express a combination of colonocyte and enterocyte phenotypes." J Cell Physiol 174(3): 362-369.
Gonzalez-Gonzalez, E., A. L. Garcia-Hernandez, R. Flores-Mejia, R. Lopez-Santiago and L. Moreno-Fierros (2015). "The protoxin Cry1Ac of Bacillus thuringiensis improves the protection conferred by intranasal immunization with Brucella abortus RB51 in a mouse model." Vet Microbiol 175(2-4): 382-388.
Gonzalez-Mariscal, L., R. Tapia and D. Chamorro (2008). "Crosstalk of tight junction components with signaling pathways." Biochim Biophys Acta 1778(3): 729-756.
Gould, F. (1998). "Sustainability of transgenic insecticidal cultivars: integrating pest genetics and ecology." Annu Rev Entomol 43: 701-726.
Guerrero, G. G., D. H. Dean and L. Moreno-Fierros (2004). "Structural implication of the induced immune response by Bacillus thuringiensis Cry proteins: role of the N-terminal region." Mol Immunol 41(12): 1177-1183.
Gupta, A., D. Nitoiu, D. Brennan-Crispi, S. Addya, N. A. Riobo, D. P. Kelsell and M. G. Mahoney (2015). "Cell cycle- and cancer-associated gene networks activated by Dsg2: evidence of cystatin A deregulation and a potential role in cell-cell adhesion." PLoS One 10(3): e0120091.
Harel, E., A. Rubinstein, A. Nissan, E. Khazanov, M. Nadler Milbauer, Y. Barenholz and B. Tirosh (2011). "Enhanced transferrin receptor expression by proinflammatory cytokines in enterocytes as a means for local delivery of drugs to inflamed gut mucosa." PLoS One 6(9): e24202.
Ibarra-Moreno, C. D., D. Ilhuicatzi-Alvarado and L. Moreno-Fierros (2021). "Differential capability of Bacillus thuringiensis Cry1Ac protoxin and toxin to induce in vivo activation of dendritic cells and B lymphocytes." Dev Comp Immunol 121: 104071.
Kalland, M. E., N. G. Oberprieler, T. Vang, K. Tasken and K. M. Torgersen (2011). "T cell-signaling network analysis reveals distinct differences between CD28 and CD2 costimulation responses in various subsets and in the MAPK pathway between resting and activated regulatory T cells." J Immunol 187(10): 5233-5245.
Khan, M. A., J. Kang and T. S. Steiner (2004). "Enteroaggregative Escherichia coli flagellin-induced interleukin-8 secretion requires Toll-like receptor 5-dependent p38 MAP kinase activation." Immunology 112(4): 651-660.
Lei, S., T. Cheng, Y. Guo, C. Li, W. Zhang and F. Zhi (2014). "Somatostatin ameliorates lipopolysaccharide-induced tight junction damage via the ERK-MAPK pathway in Caco2 cells." Eur J Cell Biol 93(7): 299-307.
Li, C., X. Bai, X. Liu, Y. Zhang, L. Liu, L. Zhang, F. Xu, Y. Yang and M. Liu (2021). "Disruption of Epithelial Barrier of Caco-2 Cell Monolayers by Excretory Secretory Products of Trichinella spiralis Might Be Related to Serine Protease." Front Microbiol 12: 634185.
Liang, N. and D. D. Kitts (2018). "Chlorogenic Acid (CGA) Isomers Alleviate Interleukin 8 (IL-8) Production in Caco-2 Cells by Decreasing Phosphorylation of p38 and Increasing Cell Integrity." Int J Mol Sci 19(12).
Meng, M., N. J. Klingensmith and C. M. Coopersmith (2017). "New insights into the gut as the driver of critical illness and organ failure." Curr Opin Crit Care 23(2): 143-148.
Mitsuyama, K., A. Toyonaga, E. Sasaki, K. Watanabe, H. Tateishi, T. Nishiyama, T. Saiki, H. Ikeda, O. Tsuruta and K. Tanikawa (1994). "IL-8 as an important chemoattractant for neutrophils in ulcerative colitis and Crohn's disease." Clin Exp Immunol 96(3): 432-436.
Moreno-Fierros, L., A. L. Garcia-Hernandez, D. Ilhuicatzi-Alvarado, L. Rivera-Santiago, M. Torres-Martinez, N. Rubio-Infante and M. Legorreta-Herrera (2013). "Cry1Ac protoxin from Bacillus thuringiensis promotes macrophage activation by upregulating CD80 and CD86 and by inducing IL-6, MCP-1 and TNF-alpha cytokines." Int Immunopharmacol 17(4): 1051-1066.
Moreno-Fierros, L., N. Garcia, R. Gutierrez, R. Lopez-Revilla and R. I. Vazquez-Padron (2000). "Intranasal, rectal and intraperitoneal immunization with protoxin Cry1Ac from Bacillus thuringiensis induces compartmentalized serum, intestinal, vaginal and pulmonary immune responses in Balb/c mice." Microbes Infect 2(8): 885-890.
Rasmussen, P. E., K. W. T. Goulding, J. R. Brown, P. R. Grace, H. H. Janzen and M. Korschens (1998). "Long-term agroecosystem experiments: assessing agricultural sustainability and global change." Science 282(5390): 893-896.
Rojas-Hernandez, S., M. A. Rodriguez-Monroy, R. Lopez-Revilla, A. A. Resendiz-Albor and L. Moreno-Fierros (2004). "Intranasal coadministration of the Cry1Ac protoxin with amoebal lysates increases protection against Naegleria fowleri meningoencephalitis." Infect Immun 72(8): 4368-4375.
Rubio-Infante, N., D. Ilhuicatzi-Alvarado, M. Torres-Martinez, J. P. Reyes-Grajeda, R. Nava-Acosta, E. Gonzalez-Gonzalez and L. Moreno-Fierros (2018). "The Macrophage Activation Induced by Bacillus thuringiensis Cry1Ac Protoxin Involves ERK1/2 and p38 Pathways and the Interaction with Cell-Surface-HSP70." J Cell Biochem 119(1): 580-598.
Samak, G., R. Gangwar, L. M. Crosby, L. P. Desai, K. Wilhelm, C. M. Waters and R. Rao (2014). "Cyclic stretch disrupts apical junctional complexes in Caco-2 cell monolayers by a JNK-2-, c-Src-, and MLCK-dependent mechanism." Am J Physiol Gastrointest Liver Physiol 306(11): G947-958.
Samak, G., D. Narayanan, J. H. Jaggar and R. Rao (2011). "CaV1.3 channels and intracellular calcium mediate osmotic stress-induced N-terminal c-Jun kinase activation and disruption of tight junctions in Caco-2 CELL MONOLAYERS." J Biol Chem 286(34): 30232-30243.
Sano, H., D. K. Hsu, L. Yu, J. R. Apgar, I. Kuwabara, T. Yamanaka, M. Hirashima and F. T. Liu (2000). "Human galectin-3 is a novel chemoattractant for monocytes and macrophages." J Immunol 165(4): 2156-2164.
Santos-Vigil, K. I., D. Ilhuicatzi-Alvarado, A. L. Garcia-Hernandez, J. S. Herrera-Garcia and L. Moreno-Fierros (2018). "Study of the allergenic potential of Bacillus thuringiensis Cry1Ac toxin following intra-gastric administration in a murine model of food-allergy." Int Immunopharmacol 61: 185-196.
Servin-Garrido, R. R., D. Ilhuicatzi-Alvarado, A. J. Jimenez-Chavez and L. Moreno-Fierros (2022). "Cry1Ac Protoxin Confers Antitumor Adjuvant Effect in a Triple-Negative Breast Cancer Mouse Model by Improving Tumor Immunity." Breast Cancer (Auckl) 16: 11782234211065154.
Sinnecker, H., K. Ramaker and A. Frey (2014). "Coating with luminal gut-constituents alters adherence of nanoparticles to intestinal epithelial cells." Beilstein J Nanotechnol 5: 2308-2315.
Szymczyk, B., W. Szczurek, S. Swiatkiewicz, K. Kwiatek, Z. Sieradzki, M. Mazur, D. Bednarek and M. Reichert (2018). "Results of a 16-week Safety Assurance Study with Rats Fed Genetically Modified Bt Maize: Effect on Growth and Health Parameters." J Vet Res 62(4): 555-561.
Torres-Martinez, M., N. Rubio-Infante, A. L. Garcia-Hernandez, R. Nava-Acosta, D. Ilhuicatzi-Alvarado and L. Moreno-Fierros (2016). "Cry1Ac toxin induces macrophage activation via ERK1/2, JNK and p38 mitogen-activated protein kinases." Int J Biochem Cell Biol 78: 106-115.
Vazquez-Padron, R. I., J. Gonzales-Cabrera, C. Garcia-Tovar, L. Neri-Bazan, R. Lopez-Revilla, M. Hernandez, L. Moreno-Fierro and G. A. de la Riva (2000). "Cry1Ac protoxin from Bacillus thuringiensis sp. kurstaki HD73 binds to surface proteins in the mouse small intestine." Biochem Biophys Res Commun 271(1): 54-58.
Vazquez, R. I., L. Moreno-Fierros, L. Neri-Bazan, G. A. De La Riva and R. Lopez-Revilla (1999). "Bacillus thuringiensis Cry1Ac protoxin is a potent systemic and mucosal adjuvant." Scand J Immunol 49(6): 578-584.
Wampler, J. L., K. P. Kim, Z. Jaradat and A. K. Bhunia (2004). "Heat shock protein 60 acts as a receptor for the Listeria adhesion protein in Caco-2 cells." Infect Immun 72(2): 931-936.
Wu, X. X., X. L. Huang, R. R. Chen, T. Li, H. J. Ye, W. Xie, Z. M. Huang and G. Z. Cao (2019). "Paeoniflorin Prevents Intestinal Barrier Disruption and Inhibits Lipopolysaccharide (LPS)-Induced Inflammation in Caco-2 Cell Monolayers." Inflammation 42(6): 2215-2225.
Zhang, M., Q. Zhuo, Y. Tian, J. Piao and X. Yang (2014). "Long-term toxicity study on transgenic rice with Cry1Ac and sck genes." Food Chem Toxicol 63: 76-83.
Zhang, X., C. Hou, S. Liu, R. Liu, X. Yin, X. Liu, H. Ma, J. Wen, R. Zhou, N. Yin, Y. Jian, S. Liu and J. Wang (2021). "Effects of transgenic Bacillus Thuringiensis maize (2A-7) on the growth and development in rats." Food Chem Toxicol 158: 112694.
Zhao, P., W. Ma, Z. Hu, L. Zang, Z. Tian and K. Zhang (2016). "Filamin A (FLNA) modulates chemosensitivity to docetaxel in triple-negative breast cancer through the MAPK/ERK pathway." Tumour Biol 37(4): 5107-5115.

No competing interests reported.

GA.png

Download PDF

Version 1

posted

You are reading this latest preprint version

Cry1Ac toxin and protoxin affect intestinal barrier function by altering zonula occludens-1 and activating mitogen-activated protein kinase pathways in a Caco-2 cell monolayer model of the human intestinal mucosa

Status:

Version 1

Abstract

Figures

Introduction

Material and methods

Recombinant Cry1Ac

Cell culture

MAPK activation

Transepithelial electrical resistance (TER) measurement

Paracellular flux assay

Confocal microscopy

Membrane protein extraction

Protein identification

Results

Cry1Ac toxin and protoxin activate ERK1/2 and p38 in Caco-2 cells

Cry1Ac toxin and protoxin induce IL-8 production

Cry1Ac increases the epithelial permeability and disrupts tight junctions

Identification of plasma membrane proteins to which Cry1Ac toxin and protoxin bind

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1