ADHESION AND ANTI-INFLAMMATORY PROPERTIES OF ENTEROCOCCI FROM HEALTHY HUMAN MICROBIOTA
TOWARDS HUMAN COLONIC EPITHELIAL CELLS UULA VAINIO
Master’s thesis
University of Eastern Finland
Department of Environmental and Biological Sciences Biology
2020
UNIVERSITY OF EASTERN FINLAND
Department of Environmental and Biological Sciences, biology
VAINIO, UULA: Adhesion and anti-inflammatory properties of enterococci from healthy human microbiota towards human colonic epithelial cells.
Master’s Thesis (40 op), 58 pp., appendices 1 October 2020
keywords: adhesion, anti-inflammatory, Enterococcus, Caco-2, HT-29 ABSTRACT
Intestinal microbiota is composed of wide range of microbes colonizing colon walls and lumen.
The metabolic potential of these microbes is vast and commensal or mutualist bacteria have numerous important functions towards host including production of extra nutrients, maintenance of immune homeostasis and protection against invading enteropathogens.
Enterococcus spp. are common bacteria of intestinal microbiota of mammals and many other animals. These bacteria are widely considered as important health-promoting probiotics, while there are also infectious strains.
Mucus-covered gut epithelial cell layer forms a barrier between body interior and the lumen of the gut. Gut epithelial cells communicate e.g. with cytokines, chemical communication molecules produced by many different cell types. IL-8 is a proinflammatory cytokine being released by gut epithelial cells for example under foreign enteropathogen invasion.
In this study, seven strains of enterococci were examined in vitro with purpose to identify those being capable to attenuate inflammatory response (excretion IL-8) of gut epithelial cells towards LPS. Caco-2 and HT-29 cell lines are used in modelling gut epithelium in vitro.
Adhesion properties of the different bacterial strains were investigated since adhesion to cell surfaces and mucus could be essential in mediating their influence on the cells nearby.
All experiments were done with Enterococcus faecalis and Enterococcus faecium isolates.
Bacteria were cultivated in MRS and YCFA medium using agar plates, and broths cultures.
Adhesion experiments towards mucus were done with bacteria from both cultivations. MRS cultivated strains were used in adhesion experiments with epithelial cell lines Caco-2 and HT- 29. Enterococci were cultivated in anaerobic conditions. Caco-2 and HT-29 cells were cultured in oxic atmosphere supplemented with 5 % CO2. Adhesion experiments towards differentiating Caco-2 cells were done with 3- and 8-day old cultures, whereas adhesion towards undifferentiating HT-29 cells was experimented only with 8-day old cultures. In attenuation experiments the inflammatory response of HT-29 cells was induced using LPS from Escherichia coli. The quantity of IL-8 in the samples was measured using ELISA assay.
The studied Enterococcus spp. strains (ETYCFA-9, -16, -23i, -23p, -24i, -25i, -25p) showed adhesiveness on mucus and both human epithelium cell lines. In mucus adhesion experiment the strains adhered strongly also to BSA which was used to block free adhesion sites of microtiter wells and therefore proved to be a poor blocking agent. Enterococci adhered well towards Caco-2 cells and adhesion was found more efficient towards cells in 8-day, than in 3- day old cultures. Enterococcus adhesion was dependent on the absolute age of the cell lines as higher passage number resulted in weaker adhesion values. Results from the attenuation experiments were inconclusive as attenuation efficiency seems to depend on culture medium of bacteria. Throughout the experiments the most adherent strains were ETYCFA-9 and- 24i and they showed better attenuation capacity when cultivated in YCFA than MRS medium, while other strains had greater attenuation efficiency when grown in MRS medium. Taken together, the results suggest that studied Enterococci might have strain and environment specific potential in attenuating LPS-induced IL-8 production in HT-29 cells.
ITÄ-SUOMEN YLIOPISTO
Ympäristö- ja biotieteiden laitos, biologia
VAINIO, UULA: Terveen ihmisen suolistosta eristettyjen enterokokkien tarttuminen ihmisen paksusuolen epiteeliin ja niiden tulehduksia ehkäisevät ominaisuudet kyseisissä epiteelisoluissa.
Pro-gradu -tutkielma (40 op), 58 s., liitteitä 1 Lokakuu 2020
avainsanat: adheesio, anti-inflammatorisuus, Enterococcus, Caco-2, HT-29 TIIVISTELMÄ
Ihmisen suoliston, erityisesti paksusuolen sisäpuolta kolonisoi suuri määrä erilaisia mikrobeja kolonisoi suolen sisäpuolta. Monet näistä mikrobeista ovat mutualistisia tai kommensaalisia ja niillä on lukuisia hyödyllisiä vaikutuksia isäntäeliöön, kuten lisäravintoaineiden tuottaminen, immuunijärjestelmän tasapainon ylläpitäminen sekä suolistopatogeeni-infektioiden ehkäiseminen. Suvun Enterococcus bakteerit ovat osana monien eläinten luonnollista suolistomikrobistoa. Nämä bakteerit mielletään usein terveyttä edistäviksi probiooteiksi, mutta enterokokkeihin kuuluu myös infektiovia, virulenttisia kantoja.
Suolessa liman peittämä epiteelikudos muodostaa erottelevan rajapinnan kehon sisäosien ja suolen ontelon välille. Epiteelisolut kommunikoivat esimerkiksi sytokiinien avulla, mitkä ovat lukuisten solutyyppien erittämiä soluvälitteisiä viestimolekyylejä. IL-8 on tulehdusreaktiota indusoiva sytokiini, jota suolen epiteelisolut erittävät esimerkiksi suolistopatogeenien tunkeutuessa suolistoon.
Tässä opinnäytetyössä tutkittiin seitsemää enterokokkikantaa in vitro ja pyrittiin tunnistamaan kannat, jotka kykenevät hillitsemään suolen epiteelisolujen LPS-indusoitua tulehdusvastetta (IL-8:n eritys). Koska bakteerien kiinnittyminen epiteelin läheisyyteen on olennaista pitkäaikaisen vaikutuksen aikaansaamiseksi, myös näiden enterokokkien tartuntakyky määritettiin suolen epiteelisoluihin sekä niitä ympäröivään limaan. Suoliston epiteeliä mallinnettiin Caco-2- ja HT-29-solulinjoilla.
Kokeissa käytetyt kannat kuuluivat lajeihin Enterococcus faecalis and Enterococcus faecium. Bakteereita kasvatettiin MRS- ja YCFA-kasvatusliuoksissa sekä vastaavilla agarmaljoilla. Kumpaakin kasvatusta käytettiin bakteerikantojen limaantarttumiskokeissa, kun taas Caco-2- ja HT-29-soluihin tarttumista tutkittiin vain MRS-kasvatetuilla kannoilla.
Enterokokit kasvatettiin anaerobisissa oloissa, Caco-2- ja HT-29-solut puolestaan hapellisissa oloissa, joissa hiilidioksidipitoisuutta oli nostettu 5 %:iin. Erilaistuvia Caco-2-soluja käytettiin tarttumiskokeissa kolmen ja kahdeksan päivän ikäisinä viljelminä, kun taas erilaistumattomia HT-29-soluja vain kahdeksan päiväisinä. Tulehdusvasteen hillintäkokeissa HT-29-solujen tulehdusreaktiota indusoitiin Escherichia coli:n LPS:llä. Näytteiden IL-8 pitoisuus määritettiin ELISA-analyysillä.
Tutkittavat Enterococcus -kannat (ETYCFA-9, -16, -23i, -23p, -24i, -25i, -25p) todettiin tartuntakykyisiksi limaan, sekä kumpaankin testatuista solulinjoista. Limaantarttumiskokeissa bakteerit tarttuivat myös voimakkaasti BSA:han, mitä käytettiin mikrotiitterilevyillä vapaiden tartuntakohtien tukkimiseen, osoittautuen huonoksi tarkoitukseensa. Enterokokit tarttuivat hyvin Caco-2-soluihin, tartuntatehokkuuden ollessa parempi kahdeksanpäiväisissä kuin kolmipäiväisissä viljelmissä. Solulinjan kasvatusikä (”passage” nro) vaikutti laskevasti tartuntakykyyn. Tulehdusvasteen hillintäkokeiden tulokset olivat keskenään ristiriitaisia, mutta viittasivat muutaman kannan mahdolliseen potentiaaliin hillitä LPS-indusoitua IL-8:n tuottoa HT-29-soluissa. Tulehduksen vaimentaminen todettiin riippuvaiseksi käytetystä bakteerien kasvatusliemestä. Kokeiden selkeästi tartuntakykyisimmät kannat olivat ETYCFA-9 ja -24i, mitkä hillitsivät tulehdusta parhaiten YCFA-kasvatettuina, muut MRS-kasvatettuina.
TABLE OF CONTENTS ABBREVIATIONS
1 INTRODUCTION ... 5
2 LITERATURE OVERVIEW ... 7
2.1 Intestinal microbiota and epithelium of humans ... 7
2.1.1 Intestinal epithelial cells and their models ... 8
2.1.1.1 Caco-2 Cell line ... 9
2.1.1.2 HT-29 Cell line ... 10
2.1.2 Cytokines and IL-8 ... 11
2.1.3 Diversity of intestinal microbiota ... 12
2.1.3.1 Genus Enterococcus ... 14
2.1.4 Function and importance of intestinal microbes ... 15
2.2 Probiotics and anti-inflammatory interactions between intestines and intestinal bacteria ... 18
2.2.1 Bacterial adhesion ... 20
3 PURPOSE OF THE STUDY ... 23
4 MATERIALS AND METHODS ... 24
4.1 Bacterial strains and cultivation conditions ... 25
4.1.1 Bacterial strains and their origin ... 25
4.1.2 Bacterial cultivation ... 25
4.1.3 Preparation of bacteria for the experiments ... 26
4.2 Culturing and maintenance of epithelial cell lines: Caco-2 and HT-29 ... 27
4.2.1 Preparation of Caco-2 and HT-29 cells for the experiments ... 28
4.3 Enterococcal adhesion to mucus, Caco-2 cells and HT-29 cells ... 29
4.3.1 Preparation of mucus for adhesion ... 29
4.3.2 Adhesion assay ... 29
4.4 Enterococcal attenuation of LPS induced IL-8 production in HT-29 cells ... 30
4.5 Statistical analysis of results ... 31
5 RESULTS ... 33
5.1 Adhesion experiments ... 33
5.1.1 Adhesion to mucus ... 33
5.1.2 Adhesion to Caco-2 cells ... 35
5.1.3 Adhesion to HT-29 cells ... 36
5.2 Attenuation experiments ... 37
6 EVALUATION OF RESULTS ... 40
6.1 Adhesion experiments ... 40
6.1.1 Adhesion to mucus ... 40
6.1.2 Adhesion to Caco-2 cells ... 43
6.1.3 Adhesion to HT-29 cells ... 44
6.2 Attenuation experiments ... 44
6.3 Further studies ... 46
7 CONCLUSIONS ... 47
ACKNOWLEDGMENTS ... 48
REFERENCES ... 49 APPENDIX 1
ABBREVIATIONS
BSA bovine serum albumin
CDI Clostridioides difficile infection
CFU colony forming unit
CPM counts per minute
DNA deoxyribonucleic acid
DPBS Dulbecco’s phosphate buffered saline
ECM extracellular matrix
ELISA enzyme-linked immunosorbent assay
FBS fetal bovine serum
FMT fecal microbiota transplant
GALT gut-associated lymphoid tissue
GC guanine-cytosine
GI gastrointestinal
HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
HSA human serum albumin
IBD inflammatory bowel disease
Ig immunoglobulin
IL interleukin
LPS lipopolysaccharide
MRS de Man, Rogosa and Sharpe medium
MSCRAMM microbial surface components recognizing adhesive matrix molecule
NEAA non-essential amino acids
O/N over night
OD optical density
P passage
PBS phosphate-buffered saline
RPMI Roswell park memorial institute medium
rRNA ribosomal ribonucleic acid
RT room temperature
SCFA short chain fatty acid
SD standard deviation
SDS sodium dodecyl sulphate
TEER transepithelial electric resistance
TLR toll-like receptor
TNF-α tumour necrosis factor alpha
VRE vancomycin-resistant Enterococcus
YCFA yeast extract, casitone and fatty acid medium
5 1 INTRODUCTION
Human gastrointestinal tracts contain hundreds of trillions of microorganisms which colonize the walls of large intestine (Power et al. 2014) while the quantity of luminal bacteria is also vast (Beisner et al.2010). The metabolic potential possessed by these microbes is wide (Blaut 2013).
The intestinal microbiota, especially commensal bacteria, has numerous important functions for the host: providing extra nutrients, contributing to the digestion of complex food components, providing a protective barrier against invading pathogenic micro-organisms and maintaining immune homeostasis (Tuohy & Scott 2015, Wang et al. 2020). Enterococcus spp.
(ex Thiercelin & Jouhaud) Schleifer & Kilpper-Bälz are common constituents of normal microbiota of humans and many other animals (Lemsaddek & Tenreiro 2012) and they are considered commonly as important health-promoting probiotics (Lebeer et al. 2010). They can also be infectious especially towards people with underlying health conditions and i.e. rapid spread of enterococci with vancomycin resistance (VRE) is a true concern (Agudelo Higuita &
Huycke 2014). Important step for bacteria to mediate beneficial effects for the host is adhering on to cells and tissues e.g. intestinal mucosa of a host. In the gastrointestinal tract, adhesion is considered an essential colonization factor for any bacteria. Their binding is affected and mediated by the surface structures and proteins of bacterial and host cells (Lebeer et al. 2010).
Ability to adhere onto surfaces and to each other (autoagglutination) is an important feature and enables biofilm formation which may be important for colonization (Linke & Goldman (ed.) 2011). Adhering probiotics and commensals have close contact to the host epithelium and they can for example act by improving epithelial integrity and stimulating immune system in the intestines (Ewaschuk et al. 2008).
Inflammation is response of the immune system to harmful stimuli and initially a defence mechanism vital to health (Chen et al. 2017). However, severe infections by Gram-negative bacteria such as sepsis can result in multi-organ dysfunctions (Patel et al 2016).
Lipopolysaccharide (LPS) is known to be a key pathogenic stimulator for the dysfunctions. LPS is major component of the outer membrane of Gram-negative bacteria and under septic circumstances circulating LPS can stimulate the innate immune system, which mediates an inflammatory response (Yücel et al 2017). Inappropriately strong immune response to invading pathogens could be attenuated by for example certain probiotic bacteria (Caruso et al. 2020).
Additionally, an enteropathogen infection can evolve to dysbiosis, an autoimmune disease associated with i.e. inflammation and decreased stability, function and diversity of gut
6
microbiota (de Oliveira 2019), which can lead to chronic inflammation of the intestines and is often associated with conditions such as inflammatory bowel diseases (IBD) and Clostridioides difficile (ex Hall & O'Toole) Lawson et al. infection (CDI). Bacteria with anti-inflammatory properties could have a great role in treating such conditions (Caruso et al. 2020).
Epithelial cell layer in the intestine functions as a barrier between the body interior and the lumen of the gut. Barrier function is further improved by a highly viscous mucus layer. Interior of small intestine is covered by partial mucus layer being thin and single-layered, whereas in large intestine the layer is thick and has doubled (Lea 2015b) which enables characteristic anaerobic condition of colon (Suprasert et al. 1987). Gut epithelial cells communicate e.g. with cytokines, chemical communication molecules produced by many different cell types. IL-8 is a proinflammatory cytokine released by gut epithelial cells for example under foreign enteropathogen invasion (Fitzgerald et al. 2001). Gut epithelial cell lines used in laboratory research are often derived from types of colon carcinoma cells, thus having characteristic limitations i.e. lack of stability in culture (Lipps et al. 2013). Nonetheless, they are valuable tools with experimenting e.g. food digestion, compound bioavailability or microbe-gut interactions such as adherence, invasion and signalling with a host (Martínez-Maqueda et al.
2015). The human colonic epithelial Caco-2 cell line has been a popular in vitro model of the intestinal epithelial barrier. It has beneficial features i.e. ability to spontaneously differentiate into a cell monolayer with typical functional and morphological properties of absorptive, brush bordered enterocytes as commonly found in the small intestine (Lea 2015a). HT-29 is another cell line isolated from a primary tumour. These cells differ by an ability to express metabolic characteristics of fully mature intestinal cells e.g. enterocytes. Thus, HT-29 is widely used in bioavailability and food digest research, but also in studies of bacterial infection, invasion and adhesion (Martínez-Maqueda et al. 2015).
In this study seven strains of previously isolated Enterococcus faecalis (ex Andrewes &
Horder) Schleifer & Kilpper-Bälz and Enterococcus faecium (ex Orla-Jensen) Schleifer &
Kilpper-Bälz are examined in vitro with purpose to identify those being capable to attenuate inflammation response towards LPS in gut epithelial cells. Adhesion to cell surfaces and mucus is essential for bacteria concerning their ability to influence the cells nearby and outcompete enteropathogens. Hence, also adhesion properties of the seven strains are investigated.
7 2 LITERATURE OVERVIEW
2.1 Intestinal microbiota and epithelium of humans
Human intestines are inhabited by hundreds of trillions (>1014) of microorganisms. It is composed mostly of bacteria, but the mixture includes also archaea and some eukaryotes such as yeasts. These microbes create an intestinal microbiota which holds 500–1000 different species (Power et al. 2014, Wang et al. 2020). Intestinal microbes form internal dynamic ecosystem for the host comparable to a vital organ (Palva 2009). The metabolic potential possessed by these microbes is wide. Due to microbial metabolism, nutrients and energy otherwise out of reach of a host are being released (for example fermentation of dietary fibres and production of vitamin K). Also, microbes interfere human metabolism by modifying for example substances produced by human body or medicines into other forms (Blaut et al. 2013, Wang et al. 2020).
The gastrointestinal tract (figure 1) starts from mouth which has a great number of bacteria (e.g. Staphylococcus Rosenbach, and Streptococcus Rosenbach) due to favourable conditions for bacteria: predominant warmth, moisture and continuous nutrition supply (Kasuga et al.
1997). The stomach having gastric acid with pH<2 and being relatively oxic habitat acts as natural sterilizer since most bacteria will not survive in such conditions (Ohland & Jobin 2015).
The survivors include i.e. Lactobacillaceae and Streptococcaceae. Here colony forming units (CFU) is less than 1000/ml (Blaut 2013). In small intestine stomach acid is neutralized, bile acid is excreted, and oxygen concentration is reduced by activity of facultative anaerobes
Figure 1. Variation and numbers of gut microbiota along the gastrointestinal tract. Regional differences affect the microbial niche. Taxons shown are examples of normal local microbiota.
Edited from original figure (Kovatcheva-Datchary et al. 2013).
8
(Ohland & Jobin 2015). Environment still remains oxic because of high partial pressure of oxygen in tissues caused by active blood circulation of the gut. Oxic environment effectively disables anaerobic bacteria but aerotolerants and facultative anaerobes successfully colonize the location or can pass through to large intestine. Overall, less than 1 % of swallowed bacteria pass live through stomach to the small intestine (Blaut 2013). The epithelium of small intestine is covered by partial mucus layer, which is thin and single-layered, whereas in large intestine mucus forms two thick layers (Lea 2015b). Viscous mucus is a mixture of water, detached cells, different salts and glycoprotein complexes, so called mucins (Reese et al. 2011). In contrast to small intestine colon also has much thicker walls and slower move blocking the oxygen flow from tissues. In large intestine nearly all oxygen has been consumed and the microenvironment is anaerobic (Blaut 2013). The anaerobic condition of colon is largely dependent on healthy, thick mucus layer (Suprasert et al. 1987). This layer is necessary to reach appropriate reduction potential for microbial fermentation which concerns most importantly dietary fibres. Short chain fatty acids (SCFAs) e.g. acetate and butyrate are being produced as fermentation products which result in a decrease in pH to 5.5. pH rises back to 7 towards the end of colon (Blaut 2013).
Most nutrients are absorbed in small intestine. The numerous intestine-encircling folds, crypts, villi and microvilli have a total surface area of roughly 300 m2. That enables great nutrient absorption rates. For many herbivores, pouch-like cecum is especially important part where microbial fermenting of ingested material takes place. For others consuming lower portion of plant material in their diet (e.g. humans) the cecum is smaller, more or less abortive.
Thus, for humans, microbial fermentation happens largely in colon although the process is most active in cecum area (Reese et al. 2011). Since colon is home for an enormous number of bacteria, they have significantly larger surface area compared to the mostly flat colon walls.
Thus, all nutrients absorbed through colon can be judged as an excess of microbial need and by-products or waste of their metabolism (Salanitro et al. 1978).
2.1.1 Intestinal epithelial cells and their models
Epithelial cell layer in the intestine functions as a barrier between the body interior and the lumen of the gut. As it allows uptake of numerous nutrients, the epithelium holds at least four different mechanisms for selective transport of macromolecules across the epithelial cell layer:
passive diffusion, paracellular transport (which is controlled by tight junction structures), carrier-mediated diffusion and vesicle-mediated transcytosis. Barrier function is further
9
improved by highly viscous mucus layer (Lea 2015b). Mucus restricts direct epithelium- bacteria contact and traps secreted substances (e.g. antimicrobial peptides). The epithelial cell layer is primarily composed of four major cell types: absorptive enterocytes, goblet cells (mucus secretion), enteroendocrine (hormone secretion) and Paneth cells that secrete antimicrobial peptides (Lea 2015b).
Gut epithelium is modelled in laboratory research with carcinoma-derived cell lines that possess limitations such as culture instability (Lipps et al. 2013). Nonetheless, they are valuable tools with experimenting e.g. food digestion, compound bioavailability or microbe-gut interactions such as adherence, invasion and signalling with a host (Martínez-Maqueda et al.
2015).
2.1.1.1 Caco-2 cell line
The human colonic epithelial Caco-2 cell line is derived from a colorectal adenocarcinoma and has been a popular in vitro model of the intestinal epithelial barrier for more than 30 years (Gonzales et al. 2015). One of its most beneficial features is the ability to spontaneously differentiate into a cell monolayer with typical functional and morphological properties of absorptive, brush bordered enterocytes as commonly found in the small intestine. For example, Caco-2 cells grow to form polarized columnar epithelium and express most transporters, receptors and drug metabolizing enzymes (i.e. aminopeptidase, sulfatase) found within normal epithelium. Caco-2 cells form a polarized monolayer and complete expression of intercellular contacts after 21 days of cultivation (Lea 2015a). The differentiation process of Caco-2 cells starts after 2–4 days from full confluence, which is important to consider when executing adhesion experiments (Natoli et al. 2012). While reaching confluence the differentiating Caco- 2 cells start to polarize acquiring apical brush border with microvilli. Tight junctions are formed between adjacent cells and enzyme activities are typical to enterocytes. Due to its origin, markers of colonocytes are also present within Caco-2 cells (Engle et al. 1998).
The Caco-2 is a heterogenous cell line consisting of cells with slightly different properties which can, due to differing culture conditions, lead to growth of selected subpopulations of cells. Hence, results gained in similar experimental conditions but in different laboratories may not be directly comparable with each other. To reduce this variability also a variety of cloned Caco-2 cell lines have been established and described (Lea 2015a).
Different culturing conditions and passage numbers can also cause changes in the properties of Caco-2 cells. Rising number of passages changes the expression of typical differentiation
10
markers of enterocytes (Artursson et al. 2001), increases proliferation rate and transepithelial electric resistance (TEER) values (Briske-Andersson et al. 1997). Additionally, late passage number cells can start forming multilayers. When compared with normal intestinal epithelium Caco-2 model has limitations such as lack of mucus and lack of other cell types present in both small and large intestine. Experimenting with a single cell line one should always be cautious in extrapolating results from in vitro models to the in vivo situation (Lea 2015a).
2.1.1.2 HT-29 cell line
The HT-29 is a human colon adenocarcinoma cell line isolated from primary tumour. This cell line has an ability to express characteristics of fully mature intestinal cells e.g. enterocytes (Martínez-Maqueda et al. 2015). Hence, HT-29 is widely used in bioavailability and food digest research. Other frequent targets of study with HT-29 are intestinal immune response to bacterial infection, invasion and adhesion. In the differentiated late growth phenotype, HT-29 cells are able to form brush bordered monolayer with tight junctions (Martínez-Maqueda et al. 2015).
HT-29 can start differentiating only when being properly induced, unlike Caco-2 cells which differentiate spontaneously. Different pathways of enterocyte differentiation of HT-29 can be induced by adjusting culture conditions or by using different inducers like butyrate (Augeron
& Laboisse 1984) and acid (Fitzgerald et al. 1997). Consequently, HT-29 might be considered as a pluripotent cell line to be used in studies concerning molecular and structural functions involved in cell differentiation (Martínez-Maqueda et al. 2015).
The differentiation of HT-29 is growth-related and starts only after full confluence being reached approximately within 15 days, depending on culturing methods (Zweibaum et al.
2011). Although HT-29 cells are essentially undifferentiated, the culture is heterogenic containing few percentages (< 5%) of cells reaching form of mucus producing and columnar absorptive cells, but only after culture reaches full confluence. Subpopulations with these features can be achieved with e.g. induction by certain metabolically stressful culturing conditions (Huet G et al. 1995). For example, using methotrexate HT-29 cells were induced to differentiate into mature goblet cells, HT-29-MTX cells, which produce mucus by stable manner (Lesuffleur et al., 1990). Enzymatic activities of HT-29 are distinctly lower when compared to Caco-2 cells (e.g. lactase is totally absent) and maximum in enzyme activities (e.g.
with sucrose-isomaltase, alkaline phosphatase, aminopeptidase N) is reached only after 30 days in culture (Zweibaum et al. 2011). Nevertheless, a clear advantage of HT-29 cells over Caco-2
11
cells is their ability to produce mucin in relatively high amounts in right growth conditions (Huet et al. 1987).
HT-29 cells secrete several factors into culture medium including cytokines and growth factors promoting cell survival. When being induced properly HT-29 cells have been reported to produce a wide range of soluble pro-inflammatory substances: interleukins (including IL-1β, IL-3, IL-6, IL-8 and IL-15), cytokines such as TNF-α (tumor necrosis factor), chemokines (i.e.
fractalkine and interferon-γ-induced protein 10), pro-angiogenic factors (i.e. vascular endothelial growth factor) and immuno-modulatory cytokines such as granulocyte colony- stimulating factor. It has been suggested that similar profile of secretion can be observed also in vivo (Desai et al. 2013). HT-29 cells have also shown similar protein expression that’s characteristic to human intestinal epithelium in vivo (Lenaerts et al. 2007). In addition, expression comparison of 377 genes in HT-29 cells and other cell lines used as in vitro models of intestinal epithelium, showed that HT-29 cells (differentiated) and human colonic tissues are not significantly different (Bourgine et al. 2012).
The HT-29 cells possess also some features limiting their use. HT-29 cells consume high amounts of glucose which needs to be considered when choosing growth medium. Low glucose concentration in medium leads to inactivity of cells. For example, under glucose concentration of 25 mM the cells do not express any of the functionally typical characteristics of intestinal epithelial cells, they grow undifferentiated and unpolarized (Pinto et al. 1982). Although the HT-29 cells mimic features of small intestine enterocytes, they still are colonic cells, but they cannot be directly compared with regular colonic enterocytes since brush border-associated hydrolases are being expressed. Nor can they be directly compared with absorptive enterocytes due to lack of many relevant hydrolases. Also, the properties of ion transport are different. The closest cells to HT-29 have been premised to be human fetal colonic cells, based on the types of hydrolases expressed and intracellular glycogen concentrations (Hekmati et al. 1990).
2.1.2 Cytokines and IL-8
Cytokines are glycoproteins or soluble proteins produced by several cell types and they serve as chemical communicators between different cells. Unifying act of most cytokines is them being regulators of the inflammatory response of host while some of them also act in defence against pathogens. Most cytokines are secreted, and they bind to specific receptors on the surface of target cells. Cytokines are usually found mediating cell differentiation or growth.
Most cytokines have great potency while acting in concentration range of nanomolar to
12
femtomolar. They also act in synergy and are counter regulated by inhibitory cytokines (Fitzgerald et al. 2001).
Interleukins (IL) are a distinct group of cytokines only by name and all belong to the six families separated by their structures. IL-8 is a proinflammatory cytokine, a chemokine receptor (figure 2). It is produced by several types of cells including lymphocytes, monocytes, hepatocytes, fibroblasts and endothelial cells such as gut epithelial cells. IL-8 functions mainly as an activating factor and chemoattractant for neutrophils but also for lymphocytes and basophils. In addition, IL-8 harbours a potency to induce angiogenesis.
There are two human IL-8 receptors described: CXCR1 and CXCR2. Stimulation of such receptor by IL-8 results an immediate increase of Ca2+ ions in intracellular matrix leading to further reactions (Fitzgerald et al. 2001).
2.1.3 Diversity of intestinal microbiota
Intestines are home to a vast range of microbes and healthy human is likely to carry approximately 1.5 kg of microbes within their intestines, mostly colonizing the walls of large intestine (Palva 2009). These microbes are estimated to hold 150-fold higher number of genes than the human genome which highlights the potential for microbes to possess numerous metabolic routes foreign to human body (Power et al. 2014). Humans obtain their gut microbes, including bacteria, archaea, microscopic fungus and viruses (e.g. bacteriophages) from food and water (Albenberg & Wu 2014). Some of them stay as native species being well adapted to the environment in our gastrointestinal tract. Meanwhile others might disappear due to their poor adaptation or because of competition (Blaut 2013).
Intestines are dominated by bacteria; five bacterial phyla cover more than 95 % of all intestinal microbes: Gram-positive Firmicutes (i.e. genera Bacillus Cohn, and Enterococcus) and Actinobacteria (e.g. Bifidobacterium Orla-Jensen), Gram-negative Bacteroidetes (i.e.
genera Bacteroides Castellani & Chalmers, and Prevotella Shah & Collins) and Proteobacteria (e.g. Escherichia Castellani & Chalmers), the fifth being Verrucomicrobia (genus Akkermansia Derrien et al.). Only one significant genus from Archaea is present: Methanobrevibacter Balch and Wolfe (Kovatcheva-Datchary et al. 2013). Eukaryotic yeasts such as Candida spp.
Figure 2. Model of IL-8 highlighting the secondary structure of the molecule (Clore et al. 1990; enhanced).
13
Berkhout, are also common in the gut. Intestinal microbes are mostly fed, directly or indirectly, by undigested carbohydrates and proteins (Power et al. 2014) but also by shedding epithelial cells and mucus (Ouwerkerk et al 2013).
It is known that diet is an important determinant of human microbiota, perhaps most important since microbiota samples from humans and other omnivorous species has proved significant similarity (Ley et al. 2008).
Arumugam et al. (2011) described Enterotypes, the three prevalent and dominant groups of intestinal bacteria at subphylum level. The enterotypes are based around Prevotella, Bacteroides and Ruminococcus Sijpesteijn. In addition to being driven by species co- occurrence and composition they also reflect functional genes and abundances of certain orthologous groups (Power et al. 2014). Whether Bacteroides or Prevotella dominate the community of intestinal microbiota the so-called enterogradient is formed because commonly these genera do not co-exist in human intestine in equal proportions (Faust et al. 2012). Also, a greater proportion of Bacteroides in the human intestinal microbiota is a marker of residence in industrialized regions whereas Prevotella is associated with life in agrarian culture (Arumugam et al. 2011).
Human gut microbiota has a unique composition in each individual. Most share a small number of prevalent microbial species yet having great variation in relative abundances from one individual to another. In the study of 17 subjects more than 50 % of them shared 66 strains, 2.1 % of total identified faecal strains (Tap et al. 2009). Certain critical metabolic functions seem to be conserved across many intestinal bacteria regardless of their taxonomic relation to another (Tap et al. 2009). These functions include the machinery to transform/ferment amino acids, produce vitamins and ferment complex carbohydrates. These functional similarities have partly been identified from collective microbial genome, metagenome. Based on this idea, certain groups of “core genes” have been found from people around the world and the cores can be composed with plentiful of different taxon combinations of microbes (Power et al. 2014).
Many sub-dominant species play critical roles in optimizing the function of the intestinal microbial community. For example, hydrogen producing bacteria offer a substrate for sulphate- reducing bacteria and methanogens. Consuming hydrogen reduces colonic gas and prevents end-product inhibition of microbial fermentation (Tuohy & Scott 2015). In a study of faecal microbiota of humans and 60 other mammalian species (wild and captive) it was found that both phylogeny and diet have an influence on the bacterial diversity and richness, both increasing from carnivores to omnivores to herbivores (Ley et al. 2008).
14
Furthermore, numerous studies have associated increased richness of microbiota in taxonomic and gene level with diets favouring vegetables, fruits and other foods high in fiber (Claesson et al. 2012, Cotillard et al. 2013, Le Chatelier et al. 2013).
Composition of bacterial and archaeal strains in human microbiota have been documented to change together quite accurately. In contrast, composition of eukaryotic microbes does not show such dependency towards any of the prokaryotes (Nam et al. 2008).
2.1.3.1 Genus Enterococcus
The genus Enterococcus was first described by Thiercelin in 1899 with an article titled “Sur un diplocoque saprophyte de l'intestin susceptible de devenir pathogène, in French” (Thiercelin 1899). The name is a result from the fact of this type of bacteria being first found from feces (Pimentel et al. 2012).
Members in genus Enterococcus are common constituents of intestinal microbiota of mammals, birds and other animals. Yet they are also common elsewhere in biosphere associating with soil, surface waters, plants and food; in raw and fermented food but also in heat-treated and pasteurized products (Lemsaddek &
Tenreiro 2012). These bacteria can persist long periods of time outside the host. In the gut, enterococci are commonly considered important health-promoting probiotics. On the other hand, enterococci have also gained a reputation of being nosocomical pathogens with potency for multi- resistance against antibiotics (Huycke &
Hancock 2012).
The genus Enterococcus belongs to Firmicutes phylum and six major groups within the genus have been described (figure 3) (Pimentel et al. 2012). It is a group of
Figure 3. Phylogenetic tree of Enterococcus spp.
based on bacterial 16S rRNA gene sequences.
Relationships amongst type strains of 35 species in total are shown with groups: A: E. avium group; D? (putative): E. dispar group; E: E.
faecium group; I: E. italicus group; F: E. faecalis group; G: E. gallinarum group.
C: E. cecorum group.
Species studied in this work are underlined.
Edited from original figure (Pimentel et al. 2012)
15
Gram-positive bacteria occurring as ovoid cells in pairs, shorts chains or single. They are variable in means of cell pigmentation and flagella-mediated motility. Enterococci include members of facultative anaerobes and chemo-organotrophs. When respiration occurs production of potent oxidants rises, which has forced these bacteria to express extreme tolerance towards oxidative stress. Many species of Enterococcus can significantly increase their growth through respiration. Enterococci are devoid of cytochrome enzymes. Different enterococci exhibit very different physical properties and thus the group is very diverse on physio-chemical terms. From the early years of their discovery enterococci are known to be resistant against drying and to successfully adapt and live under harsh conditions such as high concentrations of NaCl (up to 6.5 %, w/v) and bile (up to 40 %, v/v), and also to tolerate alkaline environment well up to pH 9.6. They tolerate detergents, heavy metals, sodium hypochlorite and ethanol. In addition, they are living in a range of temperatures between 10 ºC and 45 ºC although reaching optimal growth at 35 ºC – 37 ºC, a common body temperature of their living hosts. Enterococci are able to utilize wide range of carbohydrates (e.g. N-acetylglucosamine, glucose, lactose, cellobiose, fructose, trehalose and salicin) producing acids as end products of fermentation (e.g.
lactic acid is produced during glucose fermentation). In addition, their catabolizing spectrum of energy sources includes citrate, glycerol, malate, lactate, α-keto acids and some diamino acids.
Typically, enterococci produce a cell wall -associated glycerol teichoic acid, better known as
“the streptococcal antigen D”. In laboratory Enterococcus spp. appear catalase and oxidase negative and typical GC content in their DNA (vs. total quantity of nucleotides) is around 40
% with far ends of 35.1 % and 44.9 % between species (Lemsaddek & Tenreiro 2012).
2.1.4 Function and importance of intestinal microbes
The intestinal microbiota has numerous important functions on the host: providing extra nutrients to the host, contributing the digestion of complex food components, providing a protective barrier against invading pathogenic micro-organisms, and aiding in maintaining immune homeostasis. This co-evolved symbiotic microbiota is composed with minimum of hundreds of microbial species and strains (Wang et al. 2020). The colon contents of approximately 1011 living bacterial cells per gram is found normal. Largely affected by the microbes, gut-associated lymphoid tissue (GALT) recognizes different microbes and accommodates the healthy gut microbiota keeping invading pathogens off (Tuohy & Scott 2015). When observing gnotobiotic animals (animals without intestinal microbiota) it has been easy to state that the microbes are a necessity for healthy life of highly developed animals.
16
Without microbiota the risk of infection by e.g. enteropathogens is increased by several folds (Tlaskalová-Hogenová et al. 2011). Since microbial ecosystem in intestines acts closely and has coherent functions, the whole intestinal microbiota can be rightly perceived as one functional organ (Ohland & Jobin 2015).
It is shown by many studies how the availability of food nutrients to host is dependent on intestinal microbiota (edited by De Filippo & Tuohy 2015). Many food-derived molecules (e.g.
complex plant carbohydrates) are substrates for intestinal bacteria. For example, by catabolizing these compounds (for instance, into SCFAs) they can be absorbed by the host, used as such or metabolized in liver to a form having effect elsewhere in the body (Holmes et al. 2012).
Furthermore, human genome lacks nearly all known enzymes required to degrade plant polysaccharides. First this emerged to Gill et al. 2006 when they also reported those enzymes being supplied for human by the intestinal microbiome which possesses at least 81 glycoside hydrolase families. With these enzymes, microbes are able to catabolize starch, cellulose, and unconventional sugars i.e. xylose, mannose and arabinose. Also, an important feature of healthy gut is secretion of some small-molecule sugars consumed by commensal/probiotic bacteria as their secondary source of energy, and few antimicrobial substances that restrain bacterial growth in the close proximity of the epithelium (Bollrath & Bowrie 2013).
Complementarily host offers relatively stable habitat for living and supplies the microbes with energy and nutrients through ingested foods or directly with secretions of the gastrointestinal tract, i.e. enzymes, sloughed epithelial cells, mucin and other glycoproteins (Tuohy & Scott 2015). SCFAs produced by microbes effectively regulate hosts’ immune functions (Smith et al. 2013), lipogenesis and intestinal hormone production (Samuel et al.
2008). Important vitamins produced by bacteria include group B vitamins (i.e. folate and biotin) and group K vitamins. For example, more than half of vitamin K2 needed by humans is produced by gut bacteria. It has a crucial role e.g. in preventing osteoporosis (Tan & O’Toole 2015) whereas folate acts in replication, methylation (transcriptional regulation) and repair of DNA (Pompei et al. 2006). The role of the intestinal microbiota in providing nutritionally relevant compounds for human nutrition and health is still greatly unknown. Still it is certain that these organisms do contribute digestive and metabolic functions absent from the human genome (Yatsunenko et al. 2012).
Immune system of host is closely connected to gut microbiota and interactions between them are vast. The human immune system uses microbial recognition molecules such as LPS of Gram-negative bacteria, and peptidoglycan components of both Gram-negative and Gram- positive bacteria in microbe recognition. Within the Enterobacteriaceae family, gastrointestinal
17
pathogens typically possess proteinaceous fimbria. The identifying molecules can be
“remembered” by our immune system and can then trigger appropriate inflammatory responses to maintain homeostasis (Tuohy & Scott 2015). The recognition molecules can also be directly recognized by pattern recognition receptors which then initiates immune response against them (Castaño-Rodríguez et al 2014). Homeostasis at the mucosal membrane of intestines is the result from continuous crosstalk between immune system and the Toll-like receptors (TLRs) located on the membranes of intestinal epithelial cells (Palva 2009). Intestinal microbiota responses to environmental stressors such as hosts’ age, diet and diseases with metabolic and immunological functions. Yet, environment and activity of host seem to mediate composition of intestinal bacteria more than the body itself (Tuohy & Scott 2015). Hosts’ diet is known to modify not only microbial composition but also metabolome of theirs. For instance, diet can trigger specific gene transfers to distribute utilization mechanisms within a microbial community (Hehemann et al. 2010). As an extremely complex community, behaviour of microbiota is difficult to predict and even small changes in microbiota can have an effect on the phenotype of host. Too rapid or vast change can lead into dysbiosis, which in turn increases the risk of gaining many different intestine-derived diseases and disorders e.g. IBD, allergies and intestinal cancers (Kato et al. 2014, Caruso et al. 2020).
Regardless of the benefits and necessity of microbes, there are also pathogenic activity amongst them. Common way bacteria express pathogenicity is by producing toxins. For example, specific Clostridioides species possess such activity as well as Proteus spp. Hauser.
For counterbalance some bacteria prove themselves mostly beneficial (e.g. Bifidobacterium spp.). Among them are many bacterial taxa that can be considered as commensals but, in some circumstances, activate some virulent behaviour. For example, Bacteroides spp. and Escherichia coli (ex Migula) Castellani and Chalmers, sometimes produce carcinogens while most of the time they have positive effect on the host by e.g. facilitating nutrient absorption (Nunes de Almada et al. 2015). Many pathogens (e.g. Helicobacter pylori (ex Marshall et al.) Goodwin et al. increase cytokine release initiating inflammation in intestines which can aggravate irritable bowel syndrome and accelerate cytokinesis increasing the risk of incorrectly replicated DNA. Sometimes microbes produce too much hydrogen which can be further metabolized into ammonia (NH3) or very toxic hydrogen sulphide (H2S). In large amounts hydrogen sulphide decreases utilization of SCFAs in cells, creates genotoxic radicals and initiates apoptosis. Chronic rise of hydrogen sulphide increases risk of colon cancer (Vipperla
& O’Keefe 2013).
18
2.2 Probiotics and anti-inflammatory interactions between intestines and intestinal bacteria
Probiotic actions of micro-organisms can be divided in three different levels in gut: (1) interfering the growth and survival of pathogens in the intestinal lumen, (2) improving mucosal barrier function and mucosal immune system and (3) effecting the systemic immune system and other cell-organ systems (Rijkers et al. 2010).
Living health-promoting micro-organisms have been referred as probiotics already from the 1960s whereas the initial idea of a change in the human microbiota improving health was first proposed in a paper more than a century ago (Metchnikoff 1907). Nutrients supporting probiotic growth are called prebiotics (Praznik ym. 2015). More accurately prebiotics are nondigestible substances that act as a source of energy for the gut microbiota and stimulate activity and/or growth of certain health-promoting bacteria that inhabitat human body (Capelle et al. (ed.) 2014). Modern definition of probiotics say they are “living bacteria that, when administered in adequate amounts, confer a health benefit on the host” (FAO/WHO 2001). A potentially successful probiotic strain is likely to have several beneficial properties in order to be able to exert its desirable effects for the host (Ouwehand et al. 1999). These effects include enhancing immune response (Kimura et al. 1997), prevention of gastrointestinal infections (Nemcováet al.
1998) and to showing antimutagenic and anticarcinogenic activity (Fuller & Gibson 1997).
Currently known utilized major probiotics inhabiting human intestinal tract include Lactobacillus spp. Beijerinck, Bifidobacterium spp. and Lactococcus spp. Schleifer. There are many ways for bacteria to alter microbial composition of the intestines and to induce beneficial responses in the host. They reduce pathogen survivability e.g. by lowering pH of gut lumen with acidic by-products of their metabolism (e.g. SCFAs and lactic acid) and stimulating immune defence mechanisms of the host. They also produce other compounds that limit the growth of other micro-organisms, strengthen tight junctions between gut epithelial cells and some substances that can be utilized by gut epithelial cells as a source of energy (Palva 2009, Wang et al. 2020). However, while probiotics can survive passage through gastrointestinal tract, their persistence is for the most part short term (Wang et al. 2020). Since pathogenic bacteria share some mechanisms of adhesion with commensals these have a protective potential against pathogens by competitive binding (Lebeer et al. 2010). Competition over the same surface area and nutrients for growth as the pathogens is an effective act to restrain exogenous pathogens from colonizing. Through direct contact to dendritic cells of immune system the probiotics can affect the signals regulating T cell differentiation and IgA production (Power et al. 2014).
19
Secretion of IgA by gut epithelial cells is activated only when right inducive microbes are present (Kato et al. 2014).
Having large numbers of symbionts near the intestinal epithelial surface poses great challenge to the host since it must refrain of activation of harmful immune responses to the microbes while preserving ability to launch effective responses to invading pathogens. In patients with IBD, there are errors occurring in strategies that are used by the immune system to promote the separation between the intestinal epithelium, symbiotic microorganisms and the killing of penetrating microbes. Also, a suppression of the activation of inappropriate T cell responses to resident microorganisms occurs. IBD is often accompanied with symptoms such as chronic intestinal inflammation. Prolonged inflammation during IBD or enteropathogen invasion (e.g. CDI) could be treated with bacterial therapy, proper symbiotic microorganisms with anti-inflammatory properties instead of antibiotics and immunomodulators (Caruso et al.
2020).
2.2.1 Bacterial adhesion
Adhesion is one basic property for numerous bacteria. Ability to adhere onto surfaces and to each other (autoagglutination) is an important feature towards biofilm formation, which allows bacteria to have control over their living habitat inside the colony. This is essential escape from non-optimized environment, also enabling invasion by potential pathogens (Linke & Goldman (ed.) 2011). For instance, in many cases bacterial populations are commonly found as surface- attached colonies establishing biofilm like features (Geng & Henry 2011). Together bacterial cells can also form more complex structures. Importantly bacteria can adhere also to other cells:
prokaryotic and eukaryotic cells. Different prokaryotes can form very complex biofilms.
However, when bacteria adhere to eukaryotic cells the relationship between two parties can be issued by the purpose it serves: symbiosis, commensalism or pathogenesis (Linke & Goldman (ed.) 2011).
Successful probiotic bacteria are usually colonizing the intestine at least temporarily by adhering the intestinal mucosa (Benno & Mitsuoka 1992). The adhesion of all symbiotic, commensal and pathogenic bacteria to the cells and tissues of host is considered as an essential step in mediating their effect on the host whether they prove to be beneficial or harmful. In gastrointestinal (GI) tract adhesion is a crucial colonization factor for commensal and probiotic bacteria. Surface proteins and structures of both bacterial and host cells are known to mediate binding (Lebeer et al. 2010). By adhering to the intestinal epithelial cells, mucus and
20
extracellular matrix (ECM) proteins commensal and probiotic bacteria get to close contact with the host also persisting the intestinal tract. The GI epithelium in large intestine is covered by thick and continuous layer of mucus. In contrast, small intestine has thinner and discontinuous mucus layer resulting in areas where luminal bacteria and epithelial cells can have direct contact. Also, certain conditions might reduce the mucus barrier allowing bacteria to penetrate through it and adhere the epithelial cells and ECM proteins (Johansson et al. 2013).
Additionally, differentiation stage of epithelial cells has an effect over the expression of their surface molecules (Garcia-Lorenzo et al. 2012) which further affects the bacterial adhesion to the cells (Coconnier et al. 1993).
Bacterial adhesion is very complicated process being strongly affected by environmental factors such as temperature, bacterial concentration, associated flow conditions and antibiotics, serum proteins, bacterial properties and characteristics of the material surfaces such as hydrophobicity and charge which is determined by functional groups of surface molecules (Katsikogianni & Missirlis 2004). Flow conditions are considered to have great influence on the quantity of adhered bacteria (Isberg & Barnes 2002), biofilm structure and performance (Klapper et al. 2002). Generally, high shear rate decreases the number of attached bacteria (Katsikogianni & Missirlis 2004) and biofilm forms relatively dense and thin (Chang et al.
1991). Ming et al. (1998) suggested that after initial attachment of bacteria a series of additional interactions are formed since bacterial detachment decreases with incubation time. Suspended bacteria can respond to risen shear rates by altering growth rate, metabolism, morphology and cell size (Liu & Tay 2002). Also, pH and electrolyte (e.g. NaCl, KCl) concentrations alter bacterial adhesion (McWhirter et al. 2002). Lower pH results in higher hydrophobicity of cell surface increasing adherence towards other hydrophobic surfaces, and vice versa (Bunt et al.
1993). Antibiotics tend to decrease bacterial adhesion depending on susceptibility of bacteria towards the antibiotic and the concentration of it (Schierholz et al. 2000).
Basically, bacterial adhesion to any surface consists of the initial attraction followed by absorption and finally attachment of the cells to the surface (Rijnaarts et al. 1995). Overall polymeric surface with irregularities favour bacterial adhesion and biofilm formation compared to smooth ones (Scheuerman et al. 1998). From the physical aspect adhesion process exploits many forces i.e. van der Waals attraction forces, Brownian motion, gravitational forces, hydrophobic interactions and surface electrostatic charge (Gottenbos et al. 2002), while chemotaxis (bacterial movement directed by concentration gradients of diffusible chemoattractants) and haptotaxis (surface bound chemoattractants e.g. aminoacids) contribute the process (Kirov 2003). Bacteria are nearly always charged negatively in aqueous
21
environment (Katsikogianni & Missirlis 2004). Chemotaxis occurs commonly amongst bacteria and it regulates cellular adhesion components preparing cells for interactions between other cells and surfaces (Jenal 2004). During adhesion bacterial surface polymeric structures interact with surfaces by molecular-specific reactions. The bacterial surface structures affecting the adhesion include fimbriae, pili, capsules and slime (Mack 1999). There are proteins which promote bacterial adhesion by their structure or function (e.g. fibrinogen, thrombin) and others that have inhibitory effect towards bacterial adhesion. The latter includes albumin that inhibits bacterial adhesion at least to polymer, metal and ceramic surfaces (Dickinson et al. 1997).
Surface proteins carried by bacteria for adhesion are called adhesins. Adhesin binding to specific components (often carbohydrate structures) of tissue cells or surrounding ECM is however often quite weak. Interfering these adhesion interactions serve great potential for creating treatments against bacterial infections (Pleters 2011). Bacterial adhesins are mediators of bacterial attachment to surfaces such as tissues of eukaryotic multicellular organism. To activate bacterial adhesins they are secreted all over bacterial cell envelope during which they also fold into their final conformation (van Ulsen 2011). Ubiquitous monomeric autotransporter proteins secreted by Gram-negative bacteria are an example of adhesins although they have other functions as well (Vijri 2009). Autotransporter proteins consist of a secreted passenger domain possessing the functions, C-terminal translocator domain and N-terminal signal peptide (Geng & Henry 2011). The high purpose of receptor-specific adhesion of bacteria is preventing detachment from the target surface. For example, during infection or colonization bacteria commonly adhere to host cells through specific adhesin-receptor interactions (Beachey 1981).
Both Gram-negative and -positive bacteria show a great number of protein structures called pili or fimbriae on their surfaces that are often used for adhesive means. As such these proteins are often initiating colonization (and pathogenesis). Structural study of pilins is a growing field, but the target tissues of hosts are not well described yet. A specific feature of pilins from Gram- positive bacteria seem to be their covalent intra-molecular isopeptide bonds between lysine (Lys) and asparagine (Asn) side chains. Additionally, Gram-positive bacteria have an enzyme known as sortase that anchors adhesive proteins called “microbial surface components recognizing adhesive matrix molecules” or MSCRAMMs which targets ECM proteins of host e.g. fibronectin, collagen and fibrinogen (Krishnan & Narayana 2011). For instance, a MSCRAMM called Ace is found on Enterococcus faecalis binding collagen (Rich et al. 1999).
The numerous surface proteins expressed by bacteria play an active role in different situations.
For example, enterococcal surface protein Esp has been identified as potential virulence factor but it is found to be unessential for intestinal colonization and Caco-2 cell adhesion (Heikens
22
et al. 2009). On the contrary it has important role in binding onto abiotic surfaces and formation of biofilm that usually occurs on later steps of colonization (Toledo-Arana et al. 2001).
Initial adhesion mechanisms still keep being poorly understood although collaboration between microbiologists and physicists is increasingly producing new methodological approaches to clarify the complex aspects of those mechanisms (Geng & Henry 2011).
23 3 PURPOSE OF THE STUDY
Human intestines are inhabited by a vast range of microbes possessing important potency in regulating metabolism and physiology of their host. Existing research and knowledge keep supporting the health promoting properties of commensal and probiotic microbes such as Enterococcus spp. to humans for example during foreign enteropathogen invasion. One effective way the bacteria achieve aiding the host is by attenuating intestinal inflammation for example during severe intestinal inflammation caused by pathogen-derived dysbiosis. In this study seven previously isolated Enterococcus faecalis and Enterococcus faecium strains are examined in vitro with purpose to identify those being capable to attenuate inflammation response effectively towards LPS in gut epithelial cells. Adhesion to cell surfaces and mucus is essential for bacteria concerning their ability to influence the cells nearby and outcompete enteropathogens. Hence, also adhesion properties of the seven strains are investigated.
24 4 MATERIALS AND METHODS
In this study adhesion capacity and anti-inflammatory properties of seven strains of enterococci were experimented on intestinal epithelial cells (figure 4).
Figure 4. Schematic simplifying the progress of the experiments of this study. Seven Enterococcus spp.
strains were cultured in two different media: YCFA and MRS. Epithelial cells (Caco-2, HT-29) were let to grow for 3 or 8 days. Adhesion experiments were first done between YCFA cultured bacteria and mucus. Adhesion to epithelial cells was experimented only with MRS cultured bacteria. Bacterial adhesion was tested to 3-day old and 8-day old Caco-2 cells and to 8-day old HT-29 cells. Attenuation of LPS induced IL-8 production was experimented with only HT-29 cells.
Cultivation: Cultivation:
Epithelial cell lines:
Bacterial strains:
Enterococcus faecalis / Enterococcus faecium
YCFA medium
Caco-2
MRS medium
HT-29
3-day old cells
8-day cellsold
8-day old cells
1st Experiments:
Adhesion to
mucus IL-8 attenuation
experiments Mucus
2nd Experiments:
Adhesion to epithelial cells
Final results
Measurements
&
Analysis of results
25 4.1 Bacterial strains and cultivation conditions
4.1.1 Bacterial strains and their origin
All experiments were done with E. faecalis and E. faecium. These enterococci were previously isolated from a healthy, pre-screened donor for fecal microbiota transplant (FMT) treatment (Pakola 2017). FMT is used as microbiota-balancing treatment on patients suffering recurrent C. difficile infection (CDI) (Bakken et al. 2011). Identification of the strains was based on bacterial short (500bp) 16S rRNA gene sequences (table 1). In the attenuation experiments two Bifidobacterium strains were used as positive controls: YCFA-ab ETOH 72H1 (later “Bifido 1”) and YCFA-ab ORIG 6D10a (later “Bifido 2”).
Table 1. Bacterial strains differentiated by their 16S rRNA sequences
Isolate ID Species Strain
ETYCFA-9 Enterococcus faecalis NBRC 100480 (100%) ETYCFA-16 Enterococcus faecalis NBRC 100480 (100%) ETYCFA-23i Enterococcus faecium NBRC 100486 (99%) ETYCFA-23p Enterococcus faecium ATCC 19434 (100%) ETYCFA-24i Enterococcus faecalis NBRC 100480 (100%) ETYCFA-25i Enterococcus faecalis NBRC 100480 (100%) ETYCFA-25p Enterococcus faecalis NBRC 100480 (100%) i = provisionally anti-inflammatory
p = provisionally pro-inflammatory
The attenuation of LPS-induced IL-8 production was already tested once with isolates listed above. The strains used in this study are a group of most interesting ones from a larger isolation set (Pakola 2017). At first the bacteria were cultivated in YCFA medium (appendix 1) and stored in freezer (-80 oC). The YCFA medium was produced by Satokari group in Research Program Unit, Faculty of Medicine, University of Helsinki.
4.1.2 Bacterial cultivation
The bacterial strains were cultivated in MRS (De Man, Rogosa and Sharpe) medium (LAB M™) and YCFA medium, in broth and on agar-plates (figure 5). MRS broth was prepared as manufacturer instructed: 5.5 % (w/v) of M.R.S. broth powder was mixed in mQ-water and autoclaved before use.
26 When starting a new bacterial culture, frozen stocks were melted in RT and immediately transferred to sterile and pre-warmed (room-tempered) media to 2 % suspension in sterile 15 ml centrifuge tubes (Corning®/Nunc®/Fisherbrand®). Also 2 ml cryogenic tubes (Corning®) with screw top were used in cultivations. The screw tops were let loose while cultivating to allow the anaerobe atmosphere to develop inside the tubes. After gentle mixing the tubes containing bacteria-media mixture were put inside an anaerobic jar (Anaerocult®) with a half sheet of Anaerocult A (Merck). Forming anaerobic atmosphere was proclaimed by color-changing indicator strip (Anaerotest®). The jar was placed inside an incubator (GFL 3031) at temperature of +37 oC. For each growth cycle the Enterococci were incubated for 3 days in such conditions.
Agar plates were prepared by dissolving 1.5 % (w/v) of type A agar (Sigma) to MRS broth. The solution was autoclaved and poured on Biolite tissue culture dishes (Ø 100 mm, Thermo Scientific) 20 ml on each.
4.1.3 Preparation of bacteria for the experiments
Prior each experiment the bacterial strains were cultured anaerobically from frozen stocks and rejuvenated 3 times: Melted stock was transferred in MRS/YCFA medium as 2 % dilution (60 µl of bacterial suspension in 3 ml broth) and cultured for 3 days. Resulted bacterial suspension was recultivated in fresh broth again for 3 days (as 2 % dilution). Latter step was repeated once more and bacteria from the third cultivation cycle was used in experiments. For attenuation experiments Bifido 1 and 2 were cultivated similarly as enterococci but prior experiments they were cultivated for 2 cycles, 2 days each and in YCFA medium only. Attenuation and mucus adhesion experiments were performed with bacteria cultivated in both MRS and YCFA broth, whereas only MRS cultivated bacteria were used in Caco-2 and HT-29 adhesion experiments.
Figure 5. E. faecium strain ETYCFA- 23i colonies on MRS agar plate (top) and cells observed under light microscope as a prepared suspension (bottom).
27
For adhesion experiments bacterial cells were labeled with tritium (3H) -label. [6-3H]- thymidine (PerkinElmer®, 37 MBq/ml, 0.651 TBq/mmol) was added to the cultivation media (only on 3rd cultivation cycle) as 1 % solution (10 µl/ml ≡ 0,37 MBq/ml) to label bacterial DNA with tritium marker. Tritium labeled culture was made by adding 4 % (instead of 2 %) of bacterial suspension in medium (60 µl in 1.5 ml of total culture broth) to ensure sufficient number of bacteria for adhesion experiments.
When starting an experiment, the bacterial cells were collected and washed by centrifuging the suspensions in 1.5 ml centrifuge tubes (Fisherbrand ®) for 3 minutes using speed of 10 000 rpm (9 100 g). Washing was done with Mc Coy’s 5A (experiments with HT-29 cells) or RPMI- 1640 (experiments with Caco-2 cells) medium. After washing, the pellets were resuspended into same culture medium. Optical density (OD) of each bacterial suspension was adjusted to OD600nm = 0.25 ±0.01 with the same culture medium. For mucus adhesion experiment bacterial suspensions were washed and adjusted with PBS (pH 7.4). The OD values were verified using spectrophotometer (Pharmacia Biotech, Novaspec® II) and measuring samples in semi-micro polystyrene cuvettes (VWR).
4.2 Culturing and maintenance of epithelial cell lines: Caco-2 and HT-29
Caco-2 and HT-29 cells obtained from the Leibniz Institute DSMZ (German collection of Microorganisms and Cell Cultures) were stored in liquid nitrogen (N2, -196 oC). The cells went through three passages before the first experiment. Cultivations were accomplished in vented Biolite 25 cm2 and 75 cm2 tissue culture flasks (Thermo Scientific). The cells were grown in flasks in oxic atmosphere supplemented with 5 % CO2 in an incubator (Panasonic MCO- 170AICUVH-PE) at +37 oC. Caco-2 and HT-29 cells were detached and moved to fresh medium in every 3-5 days when the confluence of the cell monolayer reached approximately 80 % of the flask basal area. The confluence was verified each time with bare eyes and when necessary with microscope (Olympus CK40). The culture state of the epithelial cells was observed within every 1-2 days (figure 6).
Caco-2 cells were cultivated in RPMI-1640 medium (incl. NaHCO₃, Sigma) with certain additives: 20 % (v/v) heat-inactivated (56 oC for 30 min.) fetal bovine serum (FBS), 1.5 % HEPES (1 M, Lonza), 1 % non-essential amino acids solution (NEAA, Lonza), 1 % L- glutamine (Ultraglutamine I, 200 mM, Lonza) and 1 % penicillin+streptomycin (pest; 100 U/ml, Lonza). The other utilized cell line, HT-29, was cultivated in Mc Coy’s 5A medium (incl.
