Lacticaseibacillus rhamnosus GG, or LGG, is one of the most studied probiotic bacterium worldwide. The health benefit properties of LGG are mainly ensured by SpaCBA pili, which are polymeric proteins located at the cell surface. These pili confer the ability to interact with the surrounding environment and notably to adhere to intestinal cells mainly through interactions with mucin glycoproteins. For an expected probiotic effect, cell viability is an essential parameter. Nevertheless, the stability and the integrity of the pili is of paramount importance, from the conservation process of bacteria to the subsequent gastrointestinal digestion. One way to preserve bacterial viability is to encapsulate them within a matrix made of milk proteins where bacteria-matrix interactions occur and play an essential role. In the present study, nanoDSF and MST analyses revealed the remarkable thermal and detergent stability, respectively, of purified native SpaCBA pili and of the corresponding recombinant pilins. Crosslinking experiments have shown that SpaC can directly interact with β-lactoglobulin, the most abundant whey protein in the matrix with a dissociation constant (Kd) around 100 µM. The preservation of SpaCBA pili on protected LGG after spray drying was evidenced. Finally, the improved survival rate of protected LGG by resisting to simulated human digestion was shown compared to unprotected LGG. This study demonstrated that LGG encapsulation in a relevant matrix is mandatory to preserve bacteria viability and functionality after spray drying and to guarantee its preservation during the subsequent human digestion.
Article
Lacticaseibacillus rhamnosus GG encapsulation with milk proteins-based matrix preserves SpaCBA pili integrity after spray-drying and improves resistance to subsequent in vitro gastrointestinal digestion
https://doi.org/10.21203/rs.3.rs-2471474/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Lacticaseibacillus rhamnosus GG
probiotic
SpaCBA pili
pilins
gastrointestinal digestion
FAO/WHO (FAO/WHO, 2002) and the ISAPP (Hill et al., 2014; Marco et al., 2021) (The International Scientific Association for Probiotics and Prebiotics) defined probiotics as follows: “live microorganisms which/that, when administered in adequate amounts, confer a health benefit on the host”. Consumer’ interest towards probiotics either via specific food consuming or as dietary supplements is continuously increasing over the years (Gomand et al., 2019a). However, benefits are still debated with inherent pros, cons and remaining unknowns to clarify (Suez et al., 2019). Lacticaseibacillus rhamnosus GG (LGG), formerly Lactobacillus rhamnosus GG (Zheng et al., 2020), belongs to probiotic lactic acid bacteria and is certainly one of the most studied around the world (Capurso, 2019) for its positive actions in the gastrointestinal tract, especially within the intestine. One of the mechanism of action is related to its adhesion ability to mucus, via mucin glycoproteins, preventing the adhesion of pathogenic bacteria and the modulation of the immune system, among others (Suez et al., 2019). Such adhesion power is related to the presence of long filamentous proteins at the surface of bacteria, called SpaCBA pili. These proteins are pivotal for probing the bacterial environment and are essential in the bacterial adhesion process (Berne et al., 2018). LGG SpaCBA pili are sortase-dependant pili that are exclusive to Gram-positive bacteria, where subunits are covalently bound to each other. LGG SpaCBA pili (Kankainen et al., 2009) are made of three pilins with specific function, a basal pilin for cell-wall anchoring (SpaB), a major pilin for driving the pilus length (SpaA) and a tip pilin for adhesion (SpaC). The 3D structures of such pilins have been successfully solved (Chaurasia et al., 2016; Kant et al., 2020; Megta et al., 2020). LGG SpaCBA pili are self-organized in a classical way to form a fully functional pilus (von Ossowski, 2017; Dos Santos Morais et al., 2020a; Kant et al., 2020), but pilins structures in the pilus context still need further work, as it was initiated for Cornebacterium diptherae pili (McConnell et al., 2021) that certainly share structural features.
Numerous studies about LGG adhesion concerned the investigation of the interaction with molecules found in the gastro-intestinal tract, notably mucin, (von Ossowski et al., 2010; Lebeer et al., 2012), where bacteria can exert their probiotic effects.
Noteworthy, one major concern for the probiotic industry is the conservation of bacteria, since using fresh bacteria cultures might be tricky at the industrial scale (Huang et al., 2017). Bacteria are often dried, by freeze- or spray drying for instance, to increase shelf life and to reduce cost (Huang et al., 2017). The survival rate after drying and the subsequent storage are the main considered issues. To protect bacteria from ensuing physical stresses and to increase shelf life, a solution consists to protect them by microencapsulation inside a matrix (Agudelo et al., 2017; Broeckx et al., 2017; Guerin et al., 2017b; Kiekens et al., 2019). This matrix could be made of various biopolymers (Razavi et al., 2021), from carbohydrates to proteins and spanning by mixtures. Non-exhaustive examples of matrices are alginate/chitosan mixtures (Hao et al., 2021) or dairy product-based matrices including whey proteins (Burgain et al., 2013; Guerin et al., 2017a; Gomand et al., 2019a; Lu et al., 2021). When proper conditions of storage have been determined, other issues need to be overcome since orally administered probiotics are exposed to multiple harsh conditions when travelling through the gastrointestinal tract (Han et al., 2021). The storage matrix can also be useful to protect bacteria from the entire digestion process and should be released functional at the expected location. LGG is known to poorly grow in milk since it does not metabolize lactose and is not able to degrade caseins (Hussain et al., 2021). Therefore, milk components-based matrices are of choice to encapsulate LGG in order to control bacteria concentration and to avoid matrix internal hydrolysis by LGG. Thus, recent studies focused on the interaction of LGG with raw milk (Dos Santos Morais et al., 2021) and dairy components (Guerin et al., 2018a, 2018b; Dos Santos Morais et al., 2020b). Via AFM-based experiments, or ELISA derived techniques, it was highlighted that SpaCBA pilus is able to interact with whey proteins (Guerin et al., 2016; Gomand et al., 2018) but not with caseins (Burgain et al., 2014). More specifically, LGG interacts with β-lactoglobulin (BLG) while no interactions have been detected with other whey proteins (BSA and α-lactalbumin) (Guerin et al., 2016, 2018a). Importantly, LGG is also capable to interact with milk fat globule membrane likely through mucin glycoproteins (Guerin et al., 2018b). Concerning the location of LGG following emulsification and drying, when LGG was formulated only with caseins, bacteria were found exclusively at the surface of the particles, while LGG was found inside particles in the presence of casein/whey proteins mixtures (80/20) (Guerin et al., 2017a), likely due to interactions with BLG, the major whey protein. LGG flocs were also observed in raw milk (Dos Santos Morais et al., 2021) or whey proteins solution but not in culture media (Gomand et al., 2019b). Although alive, bacteria integrity might be affected and the presence of functional pili with a preserved integrity is questionable. In the present work, in a first part, the thermal and chemical stability of purified SpaCBA pili and the three recombinant pilins were studied. Then, SpaCBA-BLG interactions were investigated at the molecular level since they are involved in global bacteria-matrix interactions. Then, the resistance of LGG to freeze-drying and spray drying was investigated. Finally, in vitro digestion with or without a protecting matrix, derived from spray-drying was explored, and so the presence of intact and presumably functional SpaCBA pili.
1. Bacterial culture
The probiotic strain Lacticaseibacillus rhamnosus GG (ATCC53103) (LGG WT) and pili defective derivative mutant LGG ΔspaCBA:TcR (CMPG 5357, LGG ΔspaCBA) (Lebeer et al., 2012) were used in this study. Bacterial pre-cultures were prepared by adding 500 µL of glycerol stock (stored at -80°C) in 50 mL (~ 1/100e) of MRS medium (Biokar) and placed overnight at 37°C without agitation. LGG WT cultures were prepared by seeding approximately 5.4 L (3 x 2L-bottle) of fresh MRS with a pre-culture (~ 1/35e) and incubated at 37°C without agitation until the OD600nm reached ~ 1-1.2 (4–5 h). Bacteria were harvested by centrifugation (3 000 g, 10 min). The pellets were washed with PBS pH 6.8 (Sigma) and the suspended cells were centrifuged again (3 000 g, 10 min).
2. Bacterial concentration determination
Bacteria concentration was determined by the plate count method. The appropriate amount or volume was diluted in Tryptone Salt diluent followed by serial dilutions. One milliliter of each dilution was mixed with 15 mL of MRS agar medium and repeated in duplicate. Petri dishes were incubated 48h at 37°C before counting. Presented results are the average of two counting done in three independent experiments, unless stated otherwise.
3. Protein purification
a. Native pili purification
Protein extraction and purification were performed as in (Dos Santos Morais et al., 2020a) with minor modifications for scale-up purposes. Briefly, washed pellets were suspended in digestion buffer containing mutanolysin. The cell wall was digested overnight at 37°C under gentle agitation. The supernatant was collected, diluted in 25 mM Tris pH 6.8, and filtered (0.45 µm). All diafiltration steps were performed with Amicon stirred-cell (400 mL, 100 MWCO, Millipore). Buffer exchanges differed according to the following purification steps. All subsequent chromatography steps were performed with an ÄKTA Start liquid chromatography system (Cytiva). The clarified solution was loaded at 5 mL/min on a XK (16/40) column filled with 50 mL of Capto Core 700 resins (Cytiva) equilibrated with 25 mM Tris pH 6.8. The flow through was saved, diafiltered then injected onto an ion exchange chromatography (IEX) column (Q HP, Cytiva). The presence of polysaccharides and/or glycoproteins was monitored by PAS (Periodic Acid-Schiff) staining (Glycoprotein Staining Kit, Pierce).
b. Recombinant pilins production and purification
Recombinant pilins (SpaC, SpaB and SpaA) (Uniprot: A0A7M4DTX6, A0A5P5Z9C8 and A0A5P5ZAB3) optimized synthetic genes (5' XbaI and 3' XhoI) (Life technologies), devoid of the N-term signal peptide and the C-term LPXTG motif, were overexpressed in E. coli BL21 (DE3) competent cells (Thermo Scientific) using a pET303/CT − His vector (Invitrogen). The recombinant proteins were produced with a C-term his-tag downstream a 3C-protease cleavage site and residual amino acids from subcloning steps represented in italic below.
| Name |
Recombinant protein sequence |
MW (kDa) |
Number of residues |
 |
|---|---|---|---|---|
| SpaC |
M36TDN…LPHT858LEVLFQGPLEH6 |
91.3 |
840 |
119,640 |
| SpaB |
M32ATV…LPQT207LEVLFQGPLEH6 |
21.6 |
197 |
11,460 |
| SpaA |
M34DTN… LPHT304LEVLFQGPLEH6 |
31.3 |
288 |
39,880 |
Bacteria were transformed with the corresponding plasmid and the clones were selected on LB agar medium with carbenicillin (50 mg/L). One clone (or glycerol stock) was resuspended in 25 mL LB medium supplemented with ampicilin (50 mg/L) at 37°C overnight under agitation at 180 rpm. One liter of LB medium supplemented with ampicilin (50 mg/L) was seeded with 20 mL of the overnight preculture (~ 1/50e). Protein production was induced, when the OD600nm reached ~ 0.8-1, with 0.5 mM IsoPropyl β-D-1-ThioGalactopyranoside (IPTG), i.e. after ~ 4h at 37°C under agitation. Bacterial cultures were incubated at 20°C overnight under agitation, and then harvested by centrifugation (5 000 g, 10 min). Bacterial cells were broken either by sonication in lysis buffer (25 mM HEPES pH 7.5, 300 mM NaCl, 5% glycerol, 10 mM imidazole) or using BugBuster according to manufacturer’s instructions (Millipore). The suspension was supplemented with 5 U/mL of DENARASE (c-LEcta) and centrifuged to pellet cellular debris (40,000 g, 30 min, 4°C). The supernatant was filtered at 0.22 µm and injected onto a HiTrap TALON 5 mL colunm (Cytiva). The column was extensively washed until the ABS280nm signal returns to the baseline. The proteins were eluted in a single step with the same buffer as above but containing 300 mM imidazole. The fractions of interest were pooled and diafiltered against 25 mM Tris pH 6.8 (buffer A) with stirred-cell (400 mL, 10 or 30 MWCO, Millipore). The retentate was injected onto a strong anion exchange column (5 mL HiTrap Q HP, Cytiva) at 5 mL/min, previously equilibrated with 25 mM Tris pH 6.8. Then, the column was washed with buffer A until the ABS280nm return to the initial baseline. Captured molecules were eluted with a linear gradient at 2 mL/min reaching 100% 25 mM Tris pH 6.8, 1M NaCl (buffer B) in 10 CV.
c. SDS-PAGE analysis and protein quantification
Aliquots of each step of purification were collected and acetone-precipitated if required. Purification steps were analyzed by SDS-PAGE (4–20% gradient gel, Biorad), in reducing condition (50 mM dithiothreitol, DTT), stained with Coomassie Blue (Instant Blue, Abcam). Protein concentration was estimated spectrometrically (Nanodrop 2000c, Thermo Scientific). Protein purity towards other types of macromolecules was estimated considering the 260/280 ratio and visual inspection of the UV-spectrum.
4. Size-exclusion chromatography coupled to tetra detection array (SEC-TDA)
SEC-TDA experiments were performed with a HPLC pump (LC10AD, Shimadzu) coupled to an autosampler (Viscotek VE 2001, Malvern Panalytical) and a multi-detectors system recording UV, light scattering (RALS/LALS) intrinsic viscosity and refractive index signals (Viscotek TDA305, Malvern Panalytical). The HPLC SEC columns (BioSec5 2000 Å, 500 Å, 150 Å, 5 µm, 7.8 mm ID x 300 mm, void volume ~ 6 mL, total volume ~ 12.5 mL, Agilent) were equipped with a post-column nylon filter (0.22 µm). The columns were equilibrated with PBS (pH 6.8) supplemented with sodium azide 0.02%. The flow rate and the temperature were 0.5 mL/min and 30°C respectively. Data were processed with the Omnisec software (v5.12, Malvern Panalytical). The calibration procedure was done with BSA (Sigma). The refractometer was used as the concentration detector, unless otherwise stated, and the refractive index increment value (dn/dc) used to determine the molecular weight was 0.185 mL/g. Samples were filtered through a 0.22 µm filter just before injection.
5. Nano-differential scanning fluorimetry (nanoDSF)
NanoDSF experiments were performed with a Tycho.NT6 instrument (Nanotemper Technologies GmbH). Protein samples, at ~ 1 mg/mL for recombinant pilins and at ~ 0.1 mg/mL for native pili, were loaded into two capillaries (technical replicates). Fluorescence was recorded at 330 nm and 350 nm from 35°C to 95°C (5 min) at a rate of 30°C/min. After cooling down, the same samples were analyzed at least once again.
6. Protein–protein crosslinking assays
Recombinant pilins concentrated at 10 µM were incubated for 1 h at room temperature with BLG (Sigma) at 100 µM in PBS buffer (pH 6.8) with 5 mM EDC ((1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride, Sigma Aldrich). The crosslinking reaction was stopped by addition of Laemmli buffer. Controls without EDC were performed. Samples were analyzed on precast SDS-PAGE Gradient 4–20% (Biorad) stained with Coomassie Blue (InstantBlue, Abcam).
7. Microscale thermophoresis (MST)
Proteins labelling was performed either with RED-NHS 2nd generation or RED-tris-NTA 2nd generation kits following manufacturer’s instructions (NanoTemper Technologies GmbH). Labelled proteins concentration was between 5 and 25 nM while the “ligand” concentration was up to 1380 µM BLG, 463 µM SpaC, 312 µM SpaA and 35 mM SDS. MST experiments were achieved with a Monolith NT.115Pico instrument (NanoTemper technologies GmbH) operating between 22°C to 25°C. Experiments were done in PBST buffer pH 7.4 or 6.8 (Tween20 0.005%). The solutions were loaded into standard capillaries (NanoTemper Technologies). Instrument parameters were as follows: 5% or 20% LED power, medium or high MST power, and 3/20/1 laser off/on/off, fluorescence ratio: (-1.0-0.0 s)/(1.50–2.50 s) and (-1.0-0.0 s)/(0.50-5 s) for SDS experiments. Data were analyzed with the MO.Affinity Analysis software v2.3 (NanoTemper Technologies GmbH). For SDS stability experiments, data were fitted using the Hill model with a Hill coefficient fixed to 3, giving apparent EC50app reflecting the stability over SDS exposure. For other experiments, data were fitted using the law of mass action and solving its quadratic equation with fixed “target” concentration to determine the Kd.
8. Freeze-drying and spray drying
Fifteen mL of an overnight culture of LGG at DO600nm ~ 6.5 were harvested by centrifugation (3000 g, 10 min) and washed once with 15 mL of PBS and resuspended in 100 mL of a reconstituted milk solution. The solution was prepared after rehydration of micellar caseins (Prodiet 87 B, Ingredia, Arras, France) and whey proteins (Prodiet 90 S, Ingredia, Arras, France) at 15% (w/w) overnight at 4°C and filtered through ~ 100 µm. A 80/20 mixture (caseins/whey proteins) was then prepared. Prior freeze-drying, the suspension was stored at -20°C for 24 h. Freeze-drying was performed with a Beta-1-8 LSCBasic device (CHRIST) for 48 h at -53°C and 0.37 mbar. Spray drying experiments were performed with a mini spray dryer B-290 (BUCHI). The conditions were inlet temperature of 170 ± 3°C and outlet temperature of 85 ± 3°C which allowed to produced dairy powders with a water content of 6% (w/w). Bacterial concentration was determined between 30 and 150 days or 1 and 15 days after freeze-drying (FD) and spray drying (SD) respectively.
For removing the matrix in order to perform cell wall digestion (vide infra), 15 g of dried material was resuspended gently in 200 mL of TNE buffer (50 mM Tris pH 6.8, 150 mM NaCl, 20 mM EDTA) at 4°C for 1h. The suspension was centrifuged (3 000 g, 10 min) and the pellet resuspended in 100 mL of TNE. This step was repeated one additional time. The pellet was washed with 50 mL PBS and centrifuge (3 000 g, 10 min) to remove remaining traces of EDTA. The pellet was resuspended in PBS, transferred to 2 mL microtubes and centrifuged (3 000 g, 10 min) prior to cell wall digestion.
9. In vitro gastrointestinal (GI) digestion of LGG
A single-gastric-step or a two-step (gastric and intestinal) static in vitro digestion models were slightly modified from the procedure of (Brodkorb et al., 2019) and used to study the behavior of LGG during simulated digestion. The setup temperature was 37°C, pH was 5.5 for the gastric phase and 6.25 for the intestinal phase in order to mimic the GI conditions at half gastric emptying time during a meal (Amara et al., 2019). The setup consisted of a thermo-regulated vessel allowing a continuous magnetic stirring and equipped with a pH electrode.
Fifteen mL of an overnight culture of LGG at DO600nm ~ 6.5 were harvested by centrifugation (3 000 g, 10 min) and washed once with 15 mL of PBS. The pellet was suspended in 15 mL of 10 mM MES pH 5.5, 1.4 mM CaCl2 and 150 mM NaCl.
For the gastric phase, 3 mL of a freshly prepared gastric extract solution (Rabbit Gastric Extracts RGE, Lipolytech, Marseille, France) containing 25 mg RGE15 in 3 mL buffer were added to the reaction vessel to obtain final activities of gastric lipase of 20 U/mL and pepsin of 1000 U/ml. The pH was adjusted and maintained at 5.5 and the solution was incubated in the mechanically stirred vessel for 30 min.
For the intestinal phase of digestion, at t = 30 min, 11 ml of pancreatic extract /bile salts (sodium taurodeoxycholate, NaTDC) solution were added to the mixture and the pH was kept at 6.25 for 60 min more. Therefore, the whole digestion lasts 90 minutes. After adding the pancreatic extract/bile salt solution to the reaction vessel (675 mg Porcine Pancreatic Extract PPE, Sigma Aldrich, P7545: 8× USP specifications activity, and 57 mg NaTDC), the final activity of pancreatic lipase was 2 000 U/mL and that of NaTDC was 4 mM and the gastric phase was diluted 1.7-fold. At t = 90 min, the totally of the vessel was collected, immediately centrifuged (3 000 g, 10 min), washed with PBS once and stored at 4°C prior bacteria counting and cell wall digestion.
10. Cell wall digestion
Washed pellets were suspended in 350 µL of digestion buffer (50 mM Tris pH 6.8, 150 mM NaCl, 2 mM MgCl2, 20% sucrose, mutanolysin 100 U/mL). The cell wall was digested for 2 h at 37°C under gentle agitation to avoid cell lysis. Protoplasts were pelleted by centrifugation (7 200 g, 15 min) and the supernatant was acetone-precipitated (250 µL + 1 mL cold acetone) for at least 1 h at -20°C. The pellet was resuspended in 25 µL Laemmli buffer prior SDS-PAGE analysis.
11. Statistical analysis
The data were analyzed using SigmaPlot 14.0. One-way ANOVA (Shapiro-Wilk normality test and Brown-Forsythe equal variance test), followed by an all pairwise multiple comparison procedures (Holm-Sidak method), was used to analyze the data. In the figures, means ± SEM followed by a different superscript letter indicate significant difference (P < 0.05).
1. Purification and stability of SpaCBA pili and the related recombinant pilins
Native SpaCBA pili were purified thanks to the method previously developed by Dos Santos Morais et al., (2020a) including some scale-up adjustments to process up to 5.4 L of bacterial culture at one time (yielding ~ 200 µg of purified pili). The method relies mainly on multimodal chromatography and the Capto Core 700 technology (Dos Santos Morais et al., 2020a). As usual, pili purification was monitored by SDS-PAGE stained with Coomassie Blue (Fig. 1A). In addition, since the glycosylation of SpaCBA notably through SpaC is still debated, the possible presence of glycoproteins and polysaccharides during the purification was probed to determine if previous findings about SpaCBA glycosylation could be due to co-purified contaminants during pili purification using another protocol (Tytgat et al., 2016). In our hands, no glycosylation signal following PAS staining has been detected in the last lane of the SDS-PAGE analysis corresponding to purified SpaCBA pili (Figs. 1A and 1B) while PAS staining was positive in the HMW part of the gel and even in the well at the beginning of the purification and in some intermediate steps. These results indicate that SpaCBA might not be glycosylated and that previous observations (Tytgat et al., 2016; Latousakis and Juge, 2018) might be due to the independent co-purification of glycoproteins and/or polysaccharides or the co-purification through SpaC which possesses a putative lectin-like domain (Kant et al., 2020).
Purified pili were analyzed by SEC-TDA giving a MW at the top of the peak of ~ 2.15 MDa (Fig. 1C). Considering SpaA major pilin MW (~ 30 kDa, from protein sequence) and length (~ 9 nm according to the 3D structure, PDB: 5F44), pili are mainly ~ 600 nm-long in line with previous AFM observations (Dos Santos Morais et al., 2020a; Kant et al., 2020) and transmission electron microscopy observations (Kankainen et al., 2009). Two SpaA subunits in pili context might be shorter than the sum of two recombinant pilins due to structural arrangement during polymerization through LPXTG and pilins motifs. From SEC-TDA experiments (light scattering and viscosity detectors), the calculated radius of gyration and the hydrodynamic radius of gyration were of 148 and 68 nm respectively, giving a shape factor (ρ) > 2 (Rizos et al., 2003), are in line with the rod-shaped structure of the SpaCBA pili.
To decipher and better characterize the SpaCBA pili, the three related pilins (SpaC, SpaB and SpaA) were produced by heterologous expression in E. coli. These proteins bear a C-term His-tag allowing their purification by affinity chromatography followed by an IEX chromatography step. Conductivity signals at the top of the UV peak were ~ 15 mS/cm for SpaA (similarly to native SpaCBA pili) (Dos Santos Morais et al., 2020a) and ~ 11 mS/cm for SpaC. Expectedly, SpaB was not captured by the resin, unlike the remaining contaminants. Flow through fractions containing pure SpaB was concentrated afterwards. The three proteins were obtained at high purity as shown by SDS-PAGE and SEC-UV (Figs. 2A and 2B) since no other molecular species were detectable. Remarkably, SpaA and SpaC migrated as a double band (Fig. 2A) previously attributed to the presence of an internal isopeptide bond (Chaurasia et al., 2016; Kant et al., 2020). SpaB migrated as a single band (Fig. 2A) since the isopeptide bond is not present in the crystal structure and might occur in specific conditions (Kumar Megta et al., 2019). The monomeric state of SpaB was determined by SEC using globular protein as standard, giving a MWapp ~23 kDa in line with the 21.6 kDa determined from the sequence (Fig. 2B). For SpaA and SpaC, which harbored non-globular structure, SEC-LS data gave experimental MW of ~ 29 kDa and ~ 88 kDa respectively in adequacy with monomeric state (Fig. 2C). Importantly, SpaC elution was delayed, since it eluted close to the column total volume. Such a behavior was also observed with a different stationary phase (Sephacryl, not shown) and highlighted the adhesive behavior of this protein. As determined, SpaC is monomeric in solution at the studied concentration. Previous SpaC-SpaC interactions might occur at higher concentrations (Tripathi et al., 2013) and the derived zipper model might be reconsidered since SpaC is likely to be found at the tip of the pili (Kant et al., 2020) such as the others sortase-dependant pili (Krishnan, 2015; von Ossowski, 2017).
Then, the thermal stability of the proteins was analyzed with nanoDSF, a technique which monitors variation of intrinsic fluorescence upon unfolding by applying a 30°C/min ramp from 35°C to 95°C in order to determine its inflection temperature (Ti). For native SpaCBA pili, a remarkably high temperature of inflection (Ti) of 87°C was observed, reflecting the thermostability of SpaCBA pili (Fig. 3A). The same inflection temperature was obtained once the sample cooled down and heated again (Fig. 3A). Besides being highly resistant to high temperature, the thermal denaturation of SpaCBA pili is reversible. This reversibility might be due to the internal isopeptide bonds found in both CnaB domains composing SpaA major pilin. In the same way as native SpaCBA pili, the thermal stability of the pilins was studied by nanoDSF. SpaB has one (Ti) of 53°C corresponding to the unfolding of the CnaB domain that is not reversible as shown by the reheating of the protein where no phenomenon was observed anymore (Fig. 3A). SpaA has two (Ti) of 62°C and 76°C, likely one for each CnaB domain. More precisely, the first (Ti) could be attributed to the N-term CnaB domain and the second (Ti) to C-term CnaB domain, according to the number of Trp residues present within the domain (1 vs 3, respectively) and the signal intensity (Fig. 3A). Only one transition was previously observed by circular dichroism (Chaurasia et al., 2016), when probing β-sheets unfolding. Interestingly, SpaA pilin has two (Ti) below that the (Ti) of native SpaCBA pili, 62°C and 76°C. This difference may be due to the structural arrangement induced by the isopeptide bond through polymerization via the LPXTG and pilins motifs.
Indeed, SpaA CnaB domains are likely entangled within the native structure conferring higher stability. Interestingly, when heated a second time, SpaA possessed the same profile meaning that the thermal denaturation is fully reversible such as native SpaCBA pili. This observation was previously described in CD experiments (Chaurasia et al., 2016). Concerning SpaC, two (Ti) values of 64.5 and 81°C were obtained. The first transition could be attributed to vWFA (von Willebrand factor type A) domain since the thermal denaturation profile differed in this area when SpaC was analyzed in the presence of MgCl2, a cofactor found in the crystal structure (Kant et al., 2020), and the (Ti) value increased of 0.5°C (not shown). When reheated the protein profile shares common features with the first analysis but differs from ~ 60 to 85°C. Two (Ti) were still observed, one at 81°C and the second at 64°C. The similarity could be attributed to the reversible thermal denaturation of the 3 CnaB domains, while the differences could be associates with the irreversible thermal denaturation of the vWFA domain and the perturbation of the CnaA domain in the vicinity, which are known to be less stable than CnaB domains counterparts (Echelman et al., 2016).
Finally, detergents resistance was investigated using variable SDS concentrations in MST experiments (Fig. 3B). The EC50app value was ~ 2.0 mM for SpaCBA pili and 3.3 mM for SpaC. On the other hand, EC50app was 0.3 mM and 0.5 mM for SpaB and SpaA respectively. The resistance to SDS denaturation is relatively high for SpaCBA and SpaC compared to other proteins. For example superoxide dismutase resists up to at least 10 mM SDS, while human serum albumin structure is already affected at 0.1 mM (Gudiksen et al., 2006). Overall, the results show that SpaCBA pili are remarkably resistant to temperature and highly resistance to detergent. Interestingly, even though they are potentially lost from bacterial cell wall during industrial process and/or human digestion, pili structural integrity might persist in the environment afterward due to their extreme stability.
2. Probing molecular interactions with β-lactoglobulin
Lactic acid bacteria such as LGG are routinely introduced into dairy products for fermentation and as dairy supplement for its probiotic properties. The study of bacteria-matrix interactions is a key factor to understand the underlying mechanism and design relevant and efficient food matrices. Previous works showed that LGG is capable, through its SpaCBA pili, to interact with milk components such as MFGM (Guerin et al., 2018b) and whey protein BLG in single molecule force spectroscopy AFM based-experiments (Guerin et al., 2016) and ELISA-derived experiments (Gomand et al., 2018). In order to go deeper in the understanding of BLG-SpaCBA pili molecular interactions, recombinant pilins and native pili were employed. Notably, the aim was to determine which subunit of the pili is involved in the interaction. Having His-tagged recombinant proteins in our hands, His-pull down assays were performed. Unfortunately, no results could be exploited since non-specific binding of BLG to the resin was observed (data not shown). As an alternative strategy, crosslinking experiments using zero-length crosslinker EDC were performed (Fig. 4A). This crosslinking reagent generates an amide bond between Lys and Asp/Glu amino acid side chain. The potential interaction was studied with the three pilins even if the biological location of SpaB within the cell wall make it inaccessible to the environment in biological conditions. The main question was if the interaction between BLG and LGG occurred through the major pili SpaA or the tip pilin SpaC. The crosslinking experiment followed by SDS-PAGE analysis showed no differences in migration profile for SpaB-BLG and SpaA-BLG. On the other hand, a new doublet appeared for SpaC-BLG (Fig. 4A). These bands have a MWapp above 110 kDa corresponding to a 1:1 SpaC-BLG complex (Cx 1:1). A shallow doublet is even visible above and might be attributed to a 1:2 SpaC-BLG (Cx 1:2) since BLG is known to be dimeric in our experimental conditions (Fig. 4A). These data are the first evidence of a direct interaction between SpaC and BLG. In order to quantify the strength of the interaction, MST experiments were performed to determine a dissociation constant Kd reflecting the affinity between two molecules (Fig. 4B). As expected, relative high concentration of BLG was necessary to detect modifications in MST signal (> 1 mM or > 18.2 g.L− 1 for the highest concentration). MST is an easy-to-use method and is useful to quantify the strength of an interaction when the quantity of purified material is limited (especially with the pico detector setup), such as purified native pili. For instance, solely ~ 200 µL of labelled SpaCBA at 5–25 nM (considering SpaA molar extinction coefficient) was used for a single MST experiment. The same amount was needed for labelled pilins and labelled BLG. Indeed, several labelling strategies were conducted: labelling SpaC via the His-tag (non-covalent) or covalently, labelling SpaCBA pili or BLG covalently and using SpaA as a negative control. In the cases, BLG-SpaA, as expected no differences in MST signal was observed, meaning that no interaction occurred (Fig. 4). For SpaC-BLG and vice versa, clear differences in MST signal confirmed the interaction and allowed us to determine a Kd value around 80–100 µM indicating a relatively weak affinity (Figs. 4B and 4C). Moreover, these results gave evidences about SpaC binding site since the interaction occurred with BLG in monomeric and dimeric forms. It highlighted that the interface of BLG dimerization is not involved in the interaction with SpaC, as supported by crosslinking experiments, and the observation of a 1:2 SpaC-BLG complex (Fig. 4A). Noteworthy, the same Kd value was obtained for SpaCBA-BLG (Figs. 4C), giving the first experimental evidence of a direct interaction between native purified pili and BLG, and again the implication of SpaC.. Interestingly, SpaC is thought to possess a putative lectin-like domain specific for mannose or fucose residues (Kant et al., 2020). In our hand, no MST signal differences were observed for these monosaccharides (not shown). The presence of fucose and/or mannose residues within a glycan tree on a glycoprotein might be required for molecular recognition and might be similar to the interaction mode with intestinal mucin.
3. LGG viability and pili integrity after freeze-drying and spray drying
Most studies focused on LGG survival rate after drying and subsequent shelf storage (Huang et al., 2017). But few investigated the integrity of pili which is required for probiotic effects (Kiekens et al., 2019; Suez et al., 2019). Indeed, the cell viability does not guarantee avoiding bacterial shaving upon industrial processes. Before going further and studying the resistance of encapsulated LGG to human digestion, it is necessary to determine if SpaCBA pili are still present on bacteria surface after freeze-drying or spray drying, since shear forces might remove them especially for spray drying (Tripathi et al., 2012; Kiekens et al., 2019; Gomand et al., 2020). To protect LGG from physical stresses, notably shear stress, and increasing bacterial shelf-life, a previously developed milk proteins-based encapsulation matrix at 15% (or 150 g.L− 1) composed of micellar caseins (80%) and whey proteins (20%) was employed (Burgain et al., 2013; Guerin et al., 2017a). BLG proportion in whey proteins varies from 48 to 58% (Guo and Wang, 2016) yielding an effective concentration in the present matrix of ~ 14.4–17.4 g.L− 1 or ~ 792–960 µM, far above the determined Kd of ~ 80–100 µM. At this concentration, ~ 90% of BLG-SpaC complex is formed (Fig. 4B) and might be involved at least in SpaC protection. A pellet of a fresh LGG culture was resuspended in the presence of the aforementioned matrix then freeze-dried or spray-dried. Initial bacterial concentration was ~ 1.5.108 UFC.mL− 1, ((~ 1.109 UFC.g− 1) (Fig. 5A). After freeze drying, bacterial survival rate is slightly reduced about 54% (5.1.108 UFC.g− 1, Fig. 5A). After spray drying, cell concentration is high and is even higher than the initial bacterial concentration (2.6.109 UFC.g− 1). In conclusion, no statistical significant differences were observed but a high bacterial concentration after drying is still preserved. Noteworthy, the break of LGG chains upon shear forces induced by SD (Guerin et al., 2017b) might have occurred. Indeed, LGG can form chains made of up to 5–10 bacteria (Gomand et al., 2020) counting for 1 UFC on plate counting method. A shear factor of ~ 3 was previously described (Guerin et al., 2017b) to explain the apparent increase in cell concentration. Therefore, initial concentration might have been slightly underestimated but the resulting concentration of ~ 1-10.108 UFC.g− 1 is fully compatible with concentration requirement for putative probiotic effects (Corona-Hernandez et al., 2013).
Then, LGG was exposed to cell wall digestion to analyze CWPs content. A fresh culture of LGG WT and LGG ΔspaCBA without drying processes were used as control (Dos Santos Morais et al., 2020a) (Fig. 5B). For CWPs analysis, the profile corresponding to LGG WT showed a smeared band at high MW (> 260 kDa) attributed to SpaCBA pili (Dos Santos Morais et al., 2020a) which is absent with LGG ΔspaCBA (Fig. 5B). Prior CWPs analysis, it was necessary to remove the milk proteins-based matrix. To do so, FD- or SD-encapsulated LGG were rehydrated at 4°C in TNE buffer to remove micellar caseins. Newly formed micellar caseins over rehydration might also be found in the pellet fraction, where LGG is expected, and might interfere with subsequent analysis. Then, it is mandatory to get rid of them for an effective analysis of CWPs, especially SpaCBA pili. EDTA was used to destabilize the micellar structure by removing Ca2+ ions (Griffin et al., 1988) and favored monomeric forms that will be found in the supernatant fraction. Moreover, at this temperature, no bacterial growth resumption or changes in metabolic state were expected. After several washing steps, a bacterial pellet similar to unprotected LGG was observed and bacteria cell wall digestion was performed. The profiles of LGG WT after either SD or FD showed a smeared band at high MW similar to LGG WT confirming the presence of SpaCBA pili (Fig. 5B). The pili signal is slightly less intense for FD-LGG (Fig. 5B) and might be attributed to the loss in cell viability (Fig. 5A). A band at 160 kDa is visible and is likely due to higher protein amount in the corresponding lane. This later is also visible in subsequent experiment (please see below), but is not related to spray drying. Some other protein bands are present and are attributed to residual caseins and whey proteins from the matrix despite rigorous washing steps. Nevertheless, it cannot be excluded that some minor milk proteins might have co-pelleted with LGG through interactions with other CW components. These results confirmed that SpaCBA pili are still present and presumably functional on bacterial surface after freeze-drying or spray drying and strongly support the protective role of the matrix.
4. LGG viability and pili integrity after simulated human digestion
After demonstrating that SpaCBA pili were still present on LGG after freeze-drying or spray drying in the presence of milk proteins-based matrix (Cas/whey 80/20, 15% w/v), LGG was exposed to simulated human digestion. Only SD-LGG sample was further considered, since spray drying is a less expensive technology at the industrial scale (Santivarangkna et al., 2007; Hao et al., 2021) and gave the best cell viability results (Fig. 5B). Foremost, a fresh culture of LGG was subjected to an in vitro digestion procedure to test the resistance of unprotected bacteria during human digestion, and then the survival rate and the cell wall proteins profile were analyzed (Fig. 6). After digestion, bacterial concentration plummeted from ~ 1.5.109 UFC.mL− 1 to ~ 1.5.107 UFC.mL− 1, where 2-log were lost (Fig. 6A), meaning that unprotected LGG was highly sensitive to the gastrointestinal digestion. Concerning CWPs analysis (Fig. 6B), the control lane and the profile corresponding to the digestion process was clearly different. The band at HMW totally disappeared and novel proteins bands appeared from 160 down to 3.5 kDa. This indicates that SpaCBA pili, among other proteins, were altered after gastrointestinal digestion. This result reflecting bacterial integrity is in line with bacterial counting analysis. Therefore, when unprotected, LGG and its SpaCBA pili were found to be highly sensitive to gastrointestinal digestion. Then, the putative protective effect of the matrix was tested. Considering protected LGG, the bacterial concentration at the end of digestion was significantly higher than that calculated without matrix, 2.2.108vs 1.5.107 UFC.mL− 1 (Fig. 6A), corresponding to a survival rate of ~ 14 versus 1% In contrast to unprotected LGG, where 2-log were lost, bacteria viability was well preserved thanks to the matrix. In addition, SpaCBA pili were clearly visible at the top of the gel, even if they were less intense, which is likely due to cell mortality and a partial degradation of SpaCBA pili (Fig. 6). Proteins from the matrix might also be partially hydrolyzed with the action of the proteases presents in the pancreatic extract (Fig. 6B). These results demonstrated the importance of the presence of a relevant matrix to protect LGG from human digestion fluids and to preserve, at least partially, intact and presumably functional SpaCBA pili for probiotic effects.
In the present article the extreme thermal resistance of SpaCBA pili and the reversibility of the thermal unfolding as well as its resistance to SDS denaturation were highlighted. Then, for the first time a direct and relatively weak (Kd ~80–100µM) interaction between BLG and SpaCBA pili was evidenced and this interaction is mediated through the tip pilin SpaC. Thereafter, our investigations revealed that using a relevant milk proteins-based matrix contributed to LGG viability and the preservation of SpaCBA pili after spray drying. This matrix is also essential to protect LGG during the gastrointestinal digestion by maintaining a good cell viability and preserving pili integrity, both being essential for exerting specific probiotic properties.
Data Availability
Data available on request from the authors. The data that support the findings of this study are available from one corresponding author ([email protected]) upon reasonable request.
Authors contribution
Conceptualization: R.D.S.M, F.B., J.B. and C.G.; methodology: R.D.S.M. and S.A.; validation: R.D.S.M, F.B., J.B. and C.G.; formal analysis: R.D.S.M.; investigation: R.D.S.M., S.A., L.C.-B.M. and A.S.; resources: C.G.; data curation: R.D.S.M, F.B., J.B. and C.G.; writing-original draft preparation: R.D.S.M.; writing-review and editing: R.D.S.M., F.B., J.B., and C.G.; visualization: R.D.S.M.; supervision: F.B., J.B. and C.G.; project administration: F.B., J.B. and C.G.; funding acquisition: C.G. All authors have read and agreed to the published version of the manuscript.
Conflict of interest statement
The authors declare no conflicts of interests.
Acknowledgements
The authors are grateful for the financial support of the French ANR - Agence Nationale de la Recherche (ANR-18-CE21-0003). Furthermore, we acknowledge financial support from the "Impact Biomolecules" project of the "Lorraine Université d'Excellence" (Investissements d’avenir – ANR 15-004). LGG WT and LGG ΔspaCBA were kind gift of Dr. Sarah Leeber. We thank Dr. Pierre Soule and Dr Mathilde Belnou from Nanotemper Technologies for the access to Tycho instrument.
- Agudelo, J., Cano, A., González-Martínez, C., and Chiralt, A. (2017). Disaccharide incorporation to improve survival during storage of spray dried Lactobacillus rhamnosus in whey protein-maltodextrin carriers. Journal of Functional Foods 37, 416–423. doi:10.1016/j.jff.2017.08.014.
- Amara, S., Bourlieu, C., Humbert, L., Rainteau, D., and Carrière, F. (2019). Variations in gastrointestinal lipases, pH and bile acid levels with food intake, age and diseases: Possible impact on oral lipid-based drug delivery systems. Advanced Drug Delivery Reviews 142, 3–15. doi:10.1016/j.addr.2019.03.005.
- Berne, C., Ellison, C. K., Ducret, A., and Brun, Y. V. (2018). Bacterial adhesion at the single-cell level. Nat Rev Microbiol 16, 616–627. doi:10.1038/s41579-018-0057-5.
- Brodkorb, A., Egger, L., Alminger, M., Alvito, P., Assunção, R., Ballance, S., et al. (2019). INFOGEST static in vitro simulation of gastrointestinal food digestion. Nat Protoc 14, 991–1014. doi:10.1038/s41596-018-0119-1.
- Broeckx, G., Vandenheuvel, D., Henkens, T., Kiekens, S., van den Broek, M. F. L., Lebeer, S., et al. (2017). Enhancing the viability of Lactobacillus rhamnosus GG after spray drying and during storage. International Journal of Pharmaceutics 534, 35–41. doi:10.1016/j.ijpharm.2017.09.075.
- Burgain, J., Gaiani, C., Cailliez-Grimal, C., Jeandel, C., and Scher, J. (2013). Encapsulation of Lactobacillus rhamnosus GG in microparticles: Influence of casein to whey protein ratio on bacterial survival during digestion. Innovative Food Science & Emerging Technologies 19, 233–242. doi:10.1016/j.ifset.2013.04.012.
- Burgain, J., Scher, J., Lebeer, S., Vanderleyden, J., Cailliez-Grimal, C., Corgneau, M., et al. (2014). Significance of bacterial surface molecules interactions with milk proteins to enhance microencapsulation of Lactobacillus rhamnosus GG. Food Hydrocolloids 41, 60–70. doi:10.1016/j.foodhyd.2014.03.029.
- Capurso, L. (2019). Thirty Years of Lactobacillus rhamnosus GG: A Review. Journal of Clinical Gastroenterology 53, S1. doi:10.1097/MCG.0000000000001170.
- Chaurasia, P., Pratap, S., von Ossowski, I., Palva, A., and Krishnan, V. (2016). New insights about pilus formation in gut-adapted Lactobacillus rhamnosus GG from the crystal structure of the SpaA backbone-pilin subunit. Sci Rep 6, 28664. doi:10.1038/srep28664.
- Corona-Hernandez, R. I., Álvarez-Parrilla, E., Lizardi-Mendoza, J., Islas-Rubio, A. R., Rosa, L. A. de la, and Wall-Medrano, A. (2013). Structural Stability and Viability of Microencapsulated Probiotic Bacteria: A Review. Comprehensive Reviews in Food Science and Food Safety 12, 614–628. doi:10.1111/1541-4337.12030.
- Dos Santos Morais, R., El-Kirat-Chatel, S., Burgain, J., Simard, B., Barrau, S., Paris, C., et al. (2020a). A Fast, Efficient and Easy to Implement Method to Purify Bacterial Pili From Lacticaseibacillus rhamnosus GG Based on Multimodal Chromatography. Front. Microbiol. 11. doi:10.3389/fmicb.2020.609880.
- Dos Santos Morais, R., Gaiani, C., Borges, F., and Burgain, J. (2020b). “Interactions Microbe-Matrix in Dairy Products,” in Reference Module in Food Science (Elsevier). doi:10.1016/B978-0-08-100596-5.23004-7.
- Dos Santos Morais, R., Louvet, N., Borges, F., Dumas, D., Cvetkovska-Ben Mohamed, L., Barrau, S., et al. (2021). Impact of Lacticaseibacillus rhamnosus GG on the Emulsion Stability of Raw Milk. Foods 10, 991. doi:10.3390/foods10050991.
- Echelman, D. J., Alegre-Cebollada, J., Badilla, C. L., Chang, C., Ton-That, H., and Fernández, J. M. (2016). CnaA domains in bacterial pili are efficient dissipaters of large mechanical shocks. PNAS 113, 2490–2495. doi:10.1073/pnas.1522946113.
- Goldin, B. R., Gorbach, S. L., Saxelin, M., Barakat, S., Gualtieri, L., and Salminen, S. (1992). Survival ofLactobacillus species (strain GG) in human gastrointestinal tract. Digest Dis Sci 37, 121–128. doi:10.1007/BF01308354.
- Gomand, F., Borges, F., Burgain, J., Guerin, J., Revol-Junelles, A.-M., and Gaiani, C. (2019a). Food Matrix Design for Effective Lactic Acid Bacteria Delivery. Annual Review of Food Science and Technology 10, 285–310. doi:10.1146/annurev-food-032818-121140.
- Gomand, F., Borges, F., Guerin, J., El-Kirat-Chatel, S., Francius, G., Dumas, D., et al. (2019b). Adhesive Interactions Between Lactic Acid Bacteria and β-Lactoglobulin: Specificity and Impact on Bacterial Location in Whey Protein Isolate. Front. Microbiol. 10. doi:10.3389/fmicb.2019.01512.
- Gomand, F., Borges, F., Salim, D., Burgain, J., Guerin, J., and Gaiani, C. (2018). High-throughput screening approach to evaluate the adhesive properties of bacteria to milk biomolecules. Food Hydrocolloids 84, 537–544. doi:10.1016/j.foodhyd.2018.06.038.
- Gomand, F., Mitchell, W. H., Burgain, J., Petit, J., Borges, F., Spagnolie, S. E., et al. (2020). Shaving and breaking bacterial chains with a viscous flow. Soft Matter 16, 9273–9291. doi:10.1039/D0SM00292E.
- Griffin, M. C. A., Lyster, R. L. J., and Price, J. C. (1988). The disaggregation of calcium-depleted casein micelles. European Journal of Biochemistry 174, 339–343. doi:10.1111/j.1432-1033.1988.tb14103.x.
- Gudiksen, K. L., Gitlin, I., and Whitesides, G. M. (2006). Differentiation of proteins based on characteristic patterns of association and denaturation in solutions of SDS. PNAS 103, 7968–7972.
- Guerin, J., Bacharouche, J., Burgain, J., Lebeer, S., Francius, G., Borges, F., et al. (2016). Pili of Lactobacillus rhamnosus GG mediate interaction with β-lactoglobulin. Food Hydrocolloids 58, 35–41. doi:10.1016/j.foodhyd.2016.02.016.
- Guerin, J., Burgain, J., Borges, F., Bhandari, B., Desobry, S., Scher, J., et al. (2017a). Use of imaging techniques to identify efficient controlled release systems of Lactobacillus rhamnosus GG during in vitro digestion. Food Funct. 8, 1587–1598. doi:10.1039/C6FO01737A.
- Guerin, J., Burgain, J., Francius, G., El-Kirat-Chatel, S., Beaussart, A., Scher, J., et al. (2018a). Adhesion of Lactobacillus rhamnosus GG surface biomolecules to milk proteins. Food Hydrocolloids 82, 296–303. doi:10.1016/j.foodhyd.2018.04.016.
- Guerin, J., Petit, J., Burgain, J., Borges, F., Bhandari, B., Perroud, C., et al. (2017b). Lactobacillus rhamnosus GG encapsulation by spray-drying: Milk proteins clotting control to produce innovative matrices. Journal of Food Engineering 193, 10–19.
- Guerin, J., Soligot, C., Burgain, J., Huguet, M., Francius, G., El-Kirat-Chatel, S., et al. (2018b). Adhesive interactions between milk fat globule membrane and Lactobacillus rhamnosus GG inhibit bacterial attachment to Caco-2 TC7 intestinal cell. Colloids and Surfaces B: Biointerfaces 167, 44–53. doi:10.1016/j.colsurfb.2018.03.044.
- Guo, M., and Wang, G. (2016). Whey protein polymerisation and its applications in environmentally safe adhesives. International Journal of Dairy Technology 69, 481–488. doi:10.1111/1471-0307.12303.
- Han, S., Lu, Y., Xie, J., Fei, Y., Zheng, G., Wang, Z., et al. (2021). Probiotic Gastrointestinal Transit and Colonization After Oral Administration: A Long Journey. Frontiers in Cellular and Infection Microbiology 11, 102. doi:10.3389/fcimb.2021.609722.
- Hao, F., Fu, N., Ndiaye, H., Woo, M. W., Jeantet, R., and Chen, X. D. (2021). Thermotolerance, Survival, and Stability of Lactic Acid Bacteria After Spray Drying as Affected by the Increase of Growth Temperature. Food Bioprocess Technol 14, 120–132. doi:10.1007/s11947-020-02571-1.
- Hill, C., Guarner, F., Reid, G., Gibson, G. R., Merenstein, D. J., Pot, B., et al. (2014). The International Scientific Association for Probiotics and Prebiotics consensus statement on the scope and appropriate use of the term probiotic. Nat Rev Gastroenterol Hepatol 11, 506–514. doi:10.1038/nrgastro.2014.66.
- Huang, S., Vignolles, M.-L., Chen, X. D., Le Loir, Y., Jan, G., Schuck, P., et al. (2017). Spray drying of probiotics and other food-grade bacteria: A review. Trends in Food Science & Technology 63, 1–17. doi:10.1016/j.tifs.2017.02.007.
- Hussain, N., Li, R., Takala, T. M., Tariq, M., Zaidi, A. H., and Saris, P. E. J. Generation of Lactose- and Protease-Positive Probiotic Lacticaseibacillus rhamnosus GG by Conjugation with Lactococcus lactis NCDO 712. Applied and Environmental Microbiology 87, e02957-20. doi:10.1128/AEM.02957-20.
- Joint FAO/WHO working group report on drafting guidelines for the evaluation of probiotics in food Available at: https://www.who.int/foodsafety/fs_management/en/probiotic_guidelines.pdf
- Kankainen, M., Paulin, L., Tynkkynen, S., von Ossowski, I., Reunanen, J., Partanen, P., et al. (2009). Comparative genomic analysis of Lactobacillus rhamnosus GG reveals pili containing a human- mucus binding protein. Proc. Natl. Acad. Sci. U.S.A. 106, 17193–17198. doi:10.1073/pnas.0908876106.
- Kant, A., Palva, A., von Ossowski, I., and Krishnan, V. (2020). Crystal structure of lactobacillar SpaC reveals an atypical five-domain pilus tip adhesin: Exposing its substrate-binding and assembly in SpaCBA pili. Journal of Structural Biology 211, 107571. doi:10.1016/j.jsb.2020.107571.
- Kiekens, S., Vandenheuvel, D., Broeckx, G., Claes, I., Allonsius, C., Boeck, I. D., et al. (2019). Impact of spray-drying on the pili of Lactobacillus rhamnosus GG. Microbial Biotechnology 12, 849–855. doi:10.1111/1751-7915.13426.
- Krishnan, V. (2015). Pilins in gram-positive bacteria: A structural perspective. IUBMB Life 67, 533–543. doi:10.1002/iub.1400.
- Kumar Megta, A., Palva, A., von Ossowski, I., and Krishnan, V. (2019). SpaB, an atypically adhesive basal pilin from the lactobacillar SpaCBA pilus: crystallization and X-ray diffraction analysis. Acta Cryst F 75, 731–737. doi:10.1107/S2053230X19015358.
- Latousakis, D., and Juge, N. (2018). How Sweet Are Our Gut Beneficial Bacteria? A Focus on Protein Glycosylation in Lactobacillus. International Journal of Molecular Sciences 19, 136. doi:10.3390/ijms19010136.
- Lebeer, S., Claes, I., Tytgat, H. L. P., Verhoeven, T. L. A., Marien, E., von Ossowski, I., et al. (2012). Functional analysis of Lactobacillus rhamnosus GG pili in relation to adhesion and immunomodulatory interactions with intestinal epithelial cells. Appl. Environ. Microbiol. 78, 185–193. doi:10.1128/AEM.06192-11.
- Lu, W., Fu, N., Woo, M. W., and Chen, X. D. (2021). Exploring the interactions between Lactobacillus rhamnosus GG and whey protein isolate for preservation of the viability of bacteria through spray drying. Food Funct. 12, 2995–3008. doi:10.1039/D0FO02906H.
- Marco, M. L., Sanders, M. E., Gänzle, M., Arrieta, M. C., Cotter, P. D., De Vuyst, L., et al. (2021). The International Scientific Association for Probiotics and Prebiotics (ISAPP) consensus statement on fermented foods. Nat Rev Gastroenterol Hepatol 18, 196–208. doi:10.1038/s41575-020-00390-5.
- McConnell, S. A., McAllister, R. A., Amer, B. R., Mahoney, B. J., Sue, C. K., Chang, C., et al. (2021). Sortase-assembled pili in Corynebacterium diphtheriae are built using a latch mechanism. PNAS 118. doi:10.1073/pnas.2019649118.
- Megta, A. K., Pratap, S., Kant, A., Palva, A., von Ossowski, I., and Krishnan, V. (2020). Crystal structure of the atypically adhesive SpaB basal pilus subunit: Mechanistic insights about its incorporation in lactobacillar SpaCBA pili. Current Research in Structural Biology 2, 229–238. doi:10.1016/j.crstbi.2020.11.001.
- Razavi, S., Janfaza, S., Tasnim, N., Gibson, D. L., and Hoorfar, M. (2021). Microencapsulating polymers for probiotics delivery systems: Preparation, characterization, and applications. Food Hydrocolloids 120, 106882. doi:10.1016/j.foodhyd.2021.106882.
- Rizos, A. K., Spandidos, D. A., and Krambovitis, E. (2003). Light scattering characterization of synthetic MUC-1 peptides and their behavior in dilute solution. International Journal of Molecular Medicine 12, 559–563. doi:10.3892/ijmm.12.4.559.
- Santivarangkna, C., Kulozik, U., and Foerst, P. (2007). Alternative drying processes for the industrial preservation of lactic acid starter cultures. Biotechnol Prog 23, 302–315. doi:10.1021/bp060268f.
- Suez, J., Zmora, N., Segal, E., and Elinav, E. (2019). The pros, cons, and many unknowns of probiotics. Nat Med 25, 716–729. doi:10.1038/s41591-019-0439-x.
- Tripathi, P., Beaussart, A., Alsteens, D., Dupres, V., Claes, I., von Ossowski, I., et al. (2013). Adhesion and nanomechanics of pili from the probiotic Lactobacillus rhamnosus GG. ACS Nano 7, 3685–3697. doi:10.1021/nn400705u.
- Tripathi, P., Dupres, V., Beaussart, A., Lebeer, S., Claes, I. J. J., Vanderleyden, J., et al. (2012). Deciphering the Nanometer-Scale Organization and Assembly of Lactobacillus rhamnosus GG Pili Using Atomic Force Microscopy. Langmuir 28, 2211–2216. doi:10.1021/la203834d.
- Tytgat, H. L. P., Teijlingen, N. H. van, Sullan, R. M. A., Douillard, F. P., Rasinkangas, P., Messing, M., et al. (2016). Probiotic Gut Microbiota Isolate Interacts with Dendritic Cells via Glycosylated Heterotrimeric Pili. PLOS ONE 11, e0151824. doi:10.1371/journal.pone.0151824.
- von Ossowski, I. (2017). Novel Molecular Insights about Lactobacillar Sortase-Dependent Piliation. Int J Mol Sci 18. doi:10.3390/ijms18071551.
- von Ossowski, I., Reunanen, J., Satokari, R., Vesterlund, S., Kankainen, M., Huhtinen, H., et al. (2010). Mucosal adhesion properties of the probiotic Lactobacillus rhamnosus GG SpaCBA and SpaFED pilin subunits. Appl. Environ. Microbiol. 76, 2049–2057. doi:10.1128/AEM.01958-09.
- Zheng, J., Wittouck, S., Salvetti, E., Franz, C. M. A. P., Harris, H. M. B., Mattarelli, P., et al. (2020). A taxonomic note on the genus Lactobacillus: Description of 23 novel genera, emended description of the genus Lactobacillus Beijerinck 1901, and union of Lactobacillaceae and Leuconostocaceae. International Journal of Systematic and Evolutionary Microbiology, 70, 2782–2858. doi:10.1099/ijsem.0.004107.
(Not answered)
